Skip to main content

Full text of "History of U.S. Marine Corps operations in World War II"

See other formats

History of the US. Marine p>^^ 
Operations in Worid War II: 
Peari Hari)or to Guadalcanal, 

Volume I 

U.S Marine Corps 

pcN^ m 1231m 


Headquarters United' States Marine Corps 
Washington, DC 20380-0001 

16 August 1989 


Fleet Marine Force Reference Publications (FMFRP) 12-34-1 to 12-34-V, History of the U.S. Marine 
Corps Operations in World War II, Volumes I-V, are published to ensure the retention and dissem- 
ination of useful information learned in World War 11 by the U.S. Marine Corps. This information 
is not meant to be doctrinal or to be published in Fleet Marine Force manuals. FMFRPs in the 12 
Series are a special category: reprints of historical works which are not available elsewhere. 


This set of reference publications was written by the historical branch of the U.S. Marine Corps. They 
provide the reader with a detailed knowledge of U.S. Marine Corps World War II operations, tactics, 
and organization throughout the span of World War II. From Pearl Harbor to the occupation of Ja- 
pan, these volumes are an excellent study of the U.S. Marine Corps during one of its finest hours. 
Written with appropriate historical judgement and critical analysis, many lessons observed in these 
volumes are still applicable to theaters of operation throughout the Orient. These publications stand 
as valuable sources of information for all Marines involved in contemporary warfare and for those 
who have an interest in World War II and U.S. Marine Corps history. 

Reviewed and approved this date. 



Major General, U.S. Marine Corps 

Deputy Commander for Warfighting 

Marine Corps Combat Development Command 

Quantico, Virginia 



Pearl Harbor 
to Guadalcanal 


^::^ VOLUME I 





Historical Branch, G-3 Division, Headquarters, U. S. Marine Corps 

COAUnOO.VA 22134-6050 


' ■■=: 'S 9 
v. I 

Library of Congress Catalog Card No. 58-60002 


With the recent completion of our historical monograph project, the 
Marine Corps historical program entered a new phase. This book is the first 
of a projected five-volume series covering completely, and we hope definitively, 
the history of Marine operations in World War II. 

The fifteen historical monographs published over a period of eight years 
have served to spotlight the high points in this broad field. The basic research 
which underlay their preparation will be utilized again in this project. But a 
monograph by its very nature aims at a limited objective, and in its concentra- 
tion on a single battle or campaign necessarily ignores many related subjects. 
All too often it has been difficult to avoid conveying the impression that the 
specific operation under discussion was taking place in a vacuum. Thus, while 
much valuable history has been written, the story as a whole remains untold. 

This lack the present project aims to rectify. The story of individual 
battles or campaigns, now isolated between the covers of separate publications, 
will be largely rewritten and woven together in an attempt to show events in 
proper relation to each other and in correct perspective to the war as a whole. 
In addition, new material, especially from Japanese sources, which has become 
available since the writing of the monographs, will be integrated into the story. 
Only when the broad picture is available can the significance of the Marine 
Corps' contribution to the final victory in the Pacific be fairly evaluated. 

Now a word about Volume I which sketches briefly the development of the 
Marine Corps' amphibious mission from its inception and then carries the story 
of World War II through Guadalcanal. As logistical officer of the 1st Marine 
Division, I was privileged to take part in this, our first effort to strike back 
at the Japanese. Looking rearward from the vantage point of later years 
when our materiel superiority was overwhelming, it is difficult to visualize those 
lean first months in the Pacific when there was never enough of anything, and 
Allied strategy of giving top priority to Europe meant that there would not 
be for some time to come. Thus our initial offensive quickly and richly earned 
the nickname "Operation Shoestring." But the shoestring held during those 
early critical days when its holding appeared highly questionable; and when 
it did, the ultimate outcome of the war in the Pacific ceased to remain in doubt. 




This book covers Marine Corps participation through the first precarious 
year of World War II, when disaster piled on disaster and there seemed no 
way to check Japanese aggression. Advanced bases and garrisons were iso- 
lated and destroyed: Guam, Wake, and the Philippines. The sneak attack 
on Pearl Harbor, "the day that will live in infamy," seriously crippled the 
U. S. Pacific Fleet; yet that cripple rose to turn the tide of the entire war at 
Midway. Shortly thereafter the U. S. Marines launched on Guadalcanal an 
offensive which was destined to end only on the home islands of the Empire. 

The country in general, and the Marine Corps in particular, entered 
World War II in a better state of preparedness than had been the case in any 
other previous conflict. But that is a comparative term and does not merit 
mention in the same sentence with the degree of Japanese preparedness. 
What the Marine Corps did bring into the war, however, was the priceless in- 
gredient developed during the years of peace: the amphibious doctrines and 
techniques that made possible the trans-Pacific advance — and, for that matter, 
the invasion of North Africa and the European continent. 

By publishing this operational history in a durable form, it is hoped to 
make the Marine Corps record permanently available for the study of mili- 
tary personnel, the edification of the general public, and the contemplation of 
serious scholars of military history. 

This initial volume was planned and outlined by Lieutenant Colonel Harry 
W. Edwards, former Head of the Historical Branch, G-3 Division, Head- 
quarters, U. S. Marine Corps. Much of the original writing was done by 
Lieutenant Colonel Frank O. Hough, formerly Head of the Writing Section, 
Historical Branch. Three historical monographs, Lieutenant Colonel Kobert 
D. Heinl, Jr.'s The Defense of Wake and Marines at Midway^ and Major John 
L. Zimmerman's The Guadalcanal Cam-paign^ were adapted to the needs of 
this book by Major Verle E. Ludwig, who also contributed considerable origi- 
nal writing of his own. Mr. Kenneth W. Condit wrote the chapter on landing 
craft development and shared, with Colonel Charles W. Harrison and Major 
Hubard D. Kuokka, the authorship of the chapter treating the evolution of 
amphibious doctrine. The buildup of Pacific outpost garrisons, the opening 
moves of the war, and the record of Marines in the defense of the Philippines 
were written by Mr. Henry I. Shaw, Jr. The final editing was done by Colonel 
Harrison, present Head of the Historical Branch. 

A number of the leading participants in the actions described have com- 
mented on preliminary drafts of pertinent portions of this manuscript. Their 


valuable assistance is gratefully acknowledged. Special thanks are due to those 
people who read and commented on the entire volume : Lieutenant General 
Edward A. Craig, U. S. Marine Corps, Retired ; Dr. John Miller, Office of the 
Chief of Military History, Department of the Army; Captain Frederick K. 
Loomis, U. S. Navy, Naval History Division, Office of the Chief of Naval 
Operations, Department of the Navy; and Colonel Heinl, who initiated the 
original program of monographs dealing with Marine actions in World War II. 

Mrs. Edna Clem Kelley and her successor in the Administrative and Pro- 
duction Section of the Historical Branch, Miss Kay P. Sue, ably handled the 
exacting duties involved in processing the volume from first drafts through 
final printed form. The many preliminary typescripts and the painstaking 
task of typing the final manuscript for the printer were done by Mrs. Miriam 
R. Smallwood and Mrs. Billie J. Tucker. 

Most of the maps were prepared by the Reproduction Section, Marine 
Corps Schools, Quantico, Virginia. However, we are indebted to the Office 
of the Chief of Military History, Department of the Army, for permission to 
use Maps Nos. 3, 14, 15, 20, 21, and 23-27, which were originally drafted by 
its Cartographic Branch. Official Defense Department photographs have 
been used throughout the text. ^^ /O >» - 






1. Origins of a Mission 3 

2. Evolution of Modern Amphibious Warfare, 1920-1941 .... 8 

3. Development of Landing Craft 23 

4. Marine Occupation of Iceland 35 

5. The Marine Corps on the Eve of War 47 


1. Prewar Situation in the Pacific 59 

2. Japan Strikes 70 

3. The Southern Lifeline 84 


1. Wake in the Shadow of War 95 

2. The Enemy Strikes 106 

3. Wake Under Siege 121 

4. The Fall of Wake 132 

5. Conclusions 150 


1. China and Luzon 155 

2. Bataan Prelude 172 

3. The Siege and Capture of Corregidor . 184 


1. Setting the Stage: Early Naval Operations 205 

2. Japanese Plans: Toward Midway and the North Pacific . . . 214 

3. Midway Girds for Battle 216 

4. Midway Versus the Japanese, 4-5 June 1942 221 

5. Battle of the Carrier Planes, 4 June 1942 226 





L Background and Preparations 235 

2. Guadalcanal, 7-9 August 1942 254 

3. Tulagi and Gavutu-Tanambogo 263 

4. The Battle of the Tenaru 274 

5. The Battle of the Ridge 294 

6. Action Along the Matanikau 310 

7. Japanese Counteroffensive 322 

8. Critical November 341 

9. Final Period, 9 December 1942-9 February 1943 359 


A. Bibliographical Notes 375 

B. Chronology 382 

C. Marine Task Organization and Command Lists 387 

D. Marine Casualties 395 

E. First Marine Division Operation Order — Guadalcanal 396 

F. Military Map Symbols 399 

G. Guide to Abbreviations . 400 

H. Unit Commendations 404 

Index 413 


Continental Marines • 6 

Marines of Huntington's Battalion 6 

Marines in France in World War I 12 

Bandit-Hunting Patrol in Nicaragua 12 

Experimental Amphibian Tractor 25 

Early Version of Landing Craft 25 

Air Evacuation of Wounded in Nicaragua 49 

Army Light Tank Landing at New River, N. C . 49 

Pearl Harbor Attack 72 

Japanese Landing on Guam 72 

Japanese Patrol Craft Lost at Wake 135 

Japanese Naval Troops Who Took Wake 135 

Japanese on Bataan 174 

Aerial View of Corregidor Island 186 

Effect of Japanese Bombardment of Corregidor 186 

An Army B-25, One of Doolittle's Raiders 208 

Japanese Carrier Shoho 208 




Camouflaged Lookout Tower at Sand Island 222 

Japanese Cruiser Mikuma . 222 

Crude Sketch Map of Guadalcanal 246 

Equipment for the 1st Marine Division 251 

Marine Raiders and the Crew of the Submarine Argonaut 251 

The Original Henderson Field 255 

Unloading Supplies at Guadalcanal 255 

Tulagi Island 272 

Tanambogo and Gavutu Islands 272 

Marine Commanders on Guadalcanal 278 

LVT Bridge Built by Marine Engineers 282 

Solomons Natives Guide a Patrol 282 

90mm Antiaircraft Guns of the 3d Defense Battalion 296 

105mm Howitzer of the 11th Marines 296 

Raiders' Ridge 309 

Marines of the 2d Raider Battalion 309 

The Pagoda at Henderson Field 312 

Cactus Air Force Planes 312 

Five Blasted Japanese Tanks 331 

Marine Light Tanks 331 

Japanese Torpedo Plane 338 

Naval Gunfire Support 338 

37mm Guns of the Americal Division 361 

1st Division Marines Leave Guadalcanal 361 


1. Scene of Battle Map Section 

2. Japanese Capture of Guam, 10 December 1941 77 

3. South Pacific 91 

4. Defense Installations on Wake, 8-23 December 1941 97 

5. Landing on Wake Island, 23 December 1941 . . . . Map Section 

6. Situation on Wilkes Island, 0300-Dawn, 23 December 1941 .. . 145 

7. Japanese Landing on Bataan Map Section 

8. Corregidor with Inset Showing Manila Bay .... Map Section 

9. Marshalls, Gilberts, and Eastern Carolines 206 

10. Midway Islands, June 1942 Map Section 

11. Solomon Islands with Inset Showing Santa Cruz Islands 244 

12. Makin Raid, 17-18 August 1942 284 

13. Guadalcanal and Florida Islands Map Section 

14. Initial Dispositions, 7 August 1942 Map Section 

15. Landings in Tulagi Area, 7 August 1942 Map Section 

16. The Perimeter, 12 August 1942 287 


MAPS— Continued 


17. Battle of the Tenaru, 21 August 1942 289 

18. The Perimeter, 12-14 September 1942 300 

19. Edson's Ridge— First Phase, 12-13 September 1942 304 

20. Edson's Ridge— Final Phase, 13 September 1942 307 

21. Matanikau Action, 24-27 September 1942 314 

22. Matanikau Offensive, 7-9 October 1942 318 

23. October Attacks on the Perimeter Map Section 

24. Push Toward Kokumbona, 1-4 November 1942 344 

25. Koli Point, 4-9 November 1942 Map Section 

26. Battle Area, December 1942-January 1943 Map Section 

27. XIV Corps Plan — First January Offensive Map Section 

28. Capture of Kokumbona and Advance to the Poha River, 23-25 

January 1943 Map Section 

29. Final Phase, 26 January-9 February 1943 370 


Introduction to the Marine Corps 


Origins of a Mission 


In a sense, Marines may be said to have 
existed in ancient times when the Phoeni- 
cians, and subsequently the Greeks and 
Romans, placed men aboard their ships 
for the specific purpose of fighting, in 
contrast to the crews who navigated them 
and the rowers who propelled them. How- 
ever, Marines in the modern sense date to 
Seventeenth Century England where, in 
1664, a regiment of ground troops was 
raised specifically for duty with the fleet 
as well as ashore. This unit bore the some- 
what ponderous title : "Duke of York and 
Albany's Maritime Regiment of Foot." 
Over a period of many decades of ex- 
pansion and evolution, during much of 
which nobody knew for certain Avhether 
it belonged to the Army or the Navy, this 
basic unit developed into the corps known 
today as the Royal Marines. 

By the time of the American Revolution, 
the status of the British Marines had 
jelled firmly. Thus, when the American 
Colonies revolted and began setting up 
their own armed services, they modeled 
these much along the lines of the similar 
components of the mother country, these 
being the forms with which they were most 
familiar and which suited them best tem- 
peramentally. This was true of the Con- 
tinental Marines and to an even greater 
degree of the Marine Corps, reactivated 
under the Constitution in 1798. 

In the days of wooden, sail-propelled 
ships the functions of the Marines became 
well defijied. At sea they kept order and 
were responsible for internal security. In 

combat they became the ship's small-arms 
fighters: sniping from the fighting tops, 
and on deck spearheading boarding parties 
in close action or repelling enemy board- 
ers. Ashore they guarded naval installa- 
tions, both at home and abroad, and upon 
occasion fought on land beside Army com- 
ponents. Amphibious-wise, they were 
available as trained landing parties, either 
to seize positions on hostile shores, or to 
protect the lives and property of nationals 
in foreign countries. Both the British 
and U. S. Marines have seen much such 


At the time of this writing the Marine 
Corps is 181 years old, according to its 
own reckoning, though its service has not 
been continuous. Marines celebrate their 
Corps' birthday on 10 November, this being 
the date in the year 1775 when the Con- 
tinental Congress authorized the raising 
of two battalions of Marines for the Con- 
tinental service. The scanty records ex- 
tant show nothing to indicate that those 
battalions were actually raised, but many 
Marines were recruited for service on 
board the ships of the infant Navy where 
they performed creditably in all the major 
sea actions of the Revolutionary War, 
staged two important amphibious land- 
ings in the Bahamas, and ashore partici- 
pated in the Trenton-Princeton campaign 
under General Washington. 

The Continental Marines, like the Navy 
and all but a minuscule detachment of the 
Army, passed out of existence following 
the close of the Revolutionary War. How- 


ever, foreign pressures brought the Navy 
back into existence in 1798 under the re- 
cently adopted Constitution, and on 11 
July of that year the Marine Corps was 
reactivated as a separate service within 
the naval establishment. 

Since that date Marines have fought in 
every official war the United States has 
had — and scores of obscure affairs that 
lacked official blessing but in which, to 
quote the eminent Marine writer, John W. 
Thomason, Jr., "... a man can be killed 
as dead as ever a chap was in the Ar- 
gonne."^ They have served as strictly 
naval troops, both ashore and afloat, and 
participated in extended land operations 
under Army command, notably in the 
Creek- Seminole Indian Wars of the 1830's, 
the Mexican War, both World Wars, and 
in Korea. 

All over the world, Britian's Royal Ma- 
rines were seeing much the same type of 
service. For a century or more the courses 
of the two corps ran parallel, and they 
were as functionally alike as it is possible 
for any two military organizations to be. 
Individual members of these services had 
so many interests in common that, as one 
British writer put it, they had a tendency 
to "chum up" - when ships of the two na- 
tions put in to the same ports. Even the 
present U. S. Marine emblem (adopted in 
1868) derives from that of the Royal Ma- 
rines; thougli at a glance they appear en- 
tirely different, the basic motifs of both 
are the fouled anchor and globe : the East- 
ern Hemisphere for the British, the West- 
ern for the U. S. Much in common existed 

' Capt J. W. Thomason, Jr., Fix Bayonets! 
(New York: Charles Scribner's Sons, 1955 ed.), 

- LtCol M. Rose, RMA, A Short History of the 
Royal Marities ( Deal, England : Depot Royal 
Marines, 1911 ) , 22. 

at top level, as well, and over the years 
the two organizations developed a very 
close and most cordial relationship that 
exists to this day, despite the strange evo- 
lutionary divergence that set in between 

The transition of navies from sail to 
steam began evolutionary developments 
which profoundly altered the nature of 
all shipboard duties, and temporarily 
threatened both corps with extinction. 
From this the Royal Marines emerged 
burdened with a miscellany of often in- 
congruous duties never envisioned in the 
old days, and considerably emasculated by 
lack of a single mission of overriding im- 
portance. That the effect on the U. S. 
Marines was precisely the reverse resulted 
from the fundamental difference in the 
problems facing the two nations which 
required U. S. Marines to carve out a spe- 
cial mission for themselves, though they 
traveled a long, uneven road in bringing 
this to full fruition. 

The basic problem that confronted the 
early steam navies was that of obtaining 
fuel. Sail-propelled men-of-war, on 
which all naval experience and tradition 
up to that time was based, could operate 
at sea almost indefinitely, putting in only 
to replenish provisions and water, readily 
available at nearly any port of call any- 
where in the world. But sufficient coal 
to support large-scale steamship opera- 
tions could be obtained only from well 
stocked bases, and a fleet's operating 
radius thus became limited by the loca- 
tion of such bases. If an enemy lay be- 
yond that radius, the fleet might as well 
be chained to a post so far as getting at 

'LtCol R. D. Heinl, Jr., "What Happened to 
the Royal Marines?," USNI Proceedings, Febru- 
ary 1949, 169. 


him was concerned, unless the source of 
supply could be projected farther in his 

To the British Empire, on which "the 
sun never sets," this posed no serious prob- 
lem ; it had, or could build, all the bases it 
needed without leaving its own territory. 
But the United States, with few outlying 
possessions, had genuine cause for con- 
cern. In order to give the fleet signifi- 
cant operating range in the Pacific, the 
Navy in 1878 set up a coaling station in 
Samoa, and in 1887 the government con- 
cluded a treaty with Hawaii permitting 
the establishment of another at Pearl 

But the United States had no deep- 
seated interest in the Far East during this 
era, and no serious apprehension of an at- 
tack from that direction. The Navy's 
principal concern lay in the possibility 
of being obliged to enforce the Monroe 
Doctrine in the Caribbean or South At- 
lantic. As early as 1880, far-sighted 
naval officers began turning their thoughts 
toward this mission. The cost of main- 
taining permanent bases in those areas 
would have been prohibitive, so the prob- 
lem boiled down to devising a plan for 
seizing advanced bases when and where 
strategy dictated their need and develop- 
ing these as quickly as possible to with- 
stand attack. The scattered, under- 
strength U. S. Army of that era could not 
supply sufficient trained ground troops on 
the short notice necessary to make such 
operations effective, so the Navy faced 
the problem of developing ground 
troops of its own for service with the fleet."* 

It would seem, particularly with bene- 
fit of today's hindsight, that the Marine 
Corps would be the logical choice for the 
development of this mission. However, 
this was not so apparent at the time. Ma- 
rines had never participated in this type 
of operation on anything resembling the 
scale envisioned, and they comprised a 
very small unit as compared to the blue- 
jackets. One school of thought con- 
tended that the advanced base function 
should be performed entirely by Navy 
personnel under command of naval offi- 
cers, in the interests of unity and other 
considerations.'^ The controversy, strictly 
on the theoretical level, waxed warm and 
sometimes acrimonious, giving rise at 
length to one of those perennial efforts to 
eliminate the Marines altogether." 

However, the advent of the Spanish- 
American War found the Navy wholly 
unprepared to cope with the advanced 
base problem. It was the Marine Corps 
that promptly organized an expedition- 
ary battalion, including its own artillery 
component, for the seizure of Guanta- 
namo Bay, Cuba, in order to enable the 
U. S. Fleet to operate indefinitely in the 
Caribbean waters. At Key West this 
unit underwent training in minor tactics, 
basic weapons, and musketry, and then 
landed in the target area on 10 June 1898, 
ten days before the first Army troops ar- 
rived off the coast of Cuba. There the 
Marines quickly secured a beachhead and 
successfully defended it against a nu- 
merically superior enemy. 

*E. B. Potter (ed), The United States and 
World Sea Power (New York: Prentice-Hall, 
1955), 577-578, hereinafter cited as [/. S. <£ Sea 

° For detailed discussion of this controversy, 
see W. H. Russell, "The Genesis of FMF Doc- 
trine : 1879-1899," MC Gazette, April-July 1951. 

" LtCol R. D. Heinl, .Jr., "The Cat with More 
than Nine Lives," USNI Proceedings, June 1954. 


CONTINENTAL MARINILS present a stirring sight as they charge in this symbolic painting of Revolu- 
tionary fighting by H. Charles McBarron, Jr. ( USMC 304045) 

MARINES OF HUNTINGTON'S BATTALION, ^ri< troops ashore in Cuba in 1898, captured Guan- 
tanamo Bay, used thereafter as an American naval base. ( USMC 4982) 


So expeditiously and efficiently was this 
operation conducted that its contribution 
to the speedy and decisive culmination of 
the war would be difficult to evaluate. 
This also greatly strengthened tlie Ma- 
rine Corps' claim to the Navy's amphibi- 
ous mission, a claim tliat gained still fur- 
ther strengtli by Admiral Dewey's subse- 
quent statement that if a similar Marine 
component had served with his fleet at 
Manila Bay, the whole painful and pro- 
tracted Philippine Insurrection might 
have been avoided. 

The Spanish-American War signalized 
emergence of the United States as a world 
power. Possession of the Pliilippines 
caused the Navy to reappraise the whole 
Far East situation. The USS Charles- 
ton, convoying Army troops to Manila, 
paused en route to seize the Spanish is- 
land of Guam to serve as an advanced 
coaling station,' and annexation of Ha- 

waii followed shortly.^ Additional ad- 
vanced bases were established in the Phil- 
ippines themselves as soon as the situation 

This increasing consciousness of the 
Navy's widespread commitments and re- 
sponsibilities brought about the evolu- 
tionary developments which culminated 
in the early 1940's in the amphibious as- 
sault doctrines and techniques "which fi- 
nally made possible what Major General 
J. F. C. Fuller has called 'the most far- 
reaching tactical innovation of [World 
War II].'" « 

' Seizure of Guam reiiuiied no landing force. 
The Spanish governor had not learned about the 
declaration of war and mistoolj the token naval 

bombardment for a courtesy salute and hurried 
out to the Charleston to apologize for his inabil- 
ity to return it for lack of ammunition. He 
promptly surrendered the island upon being ap- 
prised of the facts. 

" Prior to the Spanish War, the question of 
the annexation of Hawaii had been under ne- 
gotiation off and on for many years between 
that government and the United States. In a 
treaty signed in 1875, Hawaii had been de- 
clared "an American sphere of influence." 

' Quoted in V. 8. & Sea Power, 587. 


Evolution of Modern 
Amphibious Warfare, 1920-1941 


The success of the Guantanamo Bay 
operation and the very real possibility 
that the United States' new position in 
world affairs might lead to repetitions of 
essentially the same situation led high- 
level naval strategists to become inter- 
ested in establishing a similar force on a 
permanent basis : a force capable of seiz- 
ing and defending advanced bases which 
the fleet could utilize in the prosecution of 
naval war in distant waters — waters con- 
ceivably much more distant than the Ca- 
ribbean. This in turn led to the setting 
up of a class in the fundamentals of ad- 
vanced base work at Newport, Rhode 
Island in 1901. During the winter of 
1902-1903 a Marine battalion engaged in 
advanced base defense exercises on the is- 
land of Culebra in the Caribbean in con- 
junction with the annual maneuvers 
of the fleet. Expeditionary services in 
Cuba and Panama prevented an immedi- 
ate follow-uj) to this early base defense 
instruction, but in 1910 a permanent ad- 
vanced base school was organized at New 
London, Connecticut. A year later it 
was moved to Philadelphia.^ 

By 1913 sufficient progress had been 
made in advanced base instruction to per- 

mit the formation of a permanent ad- 
vanced base force. Made up of two regi- 
ments, one of coast artillery, mines, 
searchlights, engineers, communicators, 
and other specialists for fixed defense, 
and the other of infantry and field ar- 
tillery for mobile defense, the advanced 
base force totalled about 1,750 officers and 
men. In January of 1914 it was rein- 
forced by a small Marine Corps aviation 
detachment and joined the fleet for ma- 
neuvers at Culebra.^ But the analogy be- 
tween advanced base training and the am- 
phibious assault techniques that emerged 
in World War II is easily overdrawn. 
Prior to World War I the primary inter- 
est was in defense of a base against enemy 
attack. There was no serious contem- 
plation of large-scale landings against 
heavily defended areas. 

This all but exclusive concern for the 
defense of bases was clearly borne out 
by the writing of Major Earl H. Ellis. 
Ellis, one of the most brilliant young 
Marine staff officers, was among the 
farsighted military thinkers who saw the 
prospect of war between the United 
States and Japan prior to World War I. 
Around 1913, he directed attention to the 

'J. A. Isely and P. A. Crowl, The U. 8. Ma- 
rines and AmphiMous War (Princeton: Prince- 
ton University Press, 1951), 21-22, liereinafter 
cited as Marines and Amphibious War. 

'Annual Report of the Major General Com- 
mandant of the Marine Corps, 19H (Washing- 
ton: <5P0, 1914). Hereinafter all annual re- 
ports from the Office of the Commandant will 
be cited as CMC AnRept (year). 



problems of a future Pacific conflict. To 
bring military force to bear against 
Japan, Ellis pointed out, the United 
States would have to project its fleet 
across the Pacific. To support these 
operations so far from home would re- 
quire a system of outlying bases. 
Hawaii, Guam, and the Philippines, 
which were the most important of these, 
we already possessed. Their defense 
would be of utmost importance and 
would constitute the primary mission 
of the Marine advanced base force. 
Ellis discussed in considerable detail 
the troops which would be required and 
the tactics they should employ. 

In addition to the bases already in the 
possession of the United States, Ellis 
foresaw the need of acquiring others held 
by Japan. To the Marine Corps would 
fall the job of assaulting the enemy-held 
territory. Although he did not discuss 
the problems involved nor take up the 
tactics to be employed, Ellis fore- 
shadowed the amphibious assault which 
was to be the primary mission of the 
Marine Corps in World War 11.^ 

The infant Advance Base Force was di- 
verted to other missions almost as soon as 
it was created. Hardly were the Culebra 
maneuvers of 1914 completed when the 
Marines were sent to Mexico for the seizure 
of Vera Cruz. The next year they went 
ashore in Haiti, and in 1916 unsettled con- 
ditions in Santo Domingo required the 

'Earl H. Ellis, "Naval Bases" (MS, n. d.). 
The date and origin of this MS and to whom 
it was addressed are obscure, but it appears 
that the work is either a lecture or a series of 
lectures with the following divisions : "1. 
Naval Bases ; Their Location, Resources and 
Security ; 2. The Denial of Bases ; 3. The Security 
of Advanced Bases and Advanced Base Opera- 
tions ; 4. The Advanced Base Force." 

landing of Marines in that country. Ex- 
peditionary service in these two Caribbean 
republics was to constitute a heavy and 
continuing drain on Marine Corps re- 
sources which might otherwise have been 
devoted to advanced base activities. 

The expansion of the Marine Corps to 
about 73,000 officers and men during 
World War I served as a temporary stimu- 
lant to the Advance Base Force. In spite 
of the demands for manpower resulting 
from the sending of an expeditionary force 
to France, the Advance Base Force was 
maintained at full strength throughout 
the war. By the Armistice it numbered 
6,297 officers and men.* 


Marines returning from overseas late in 
1919 picked up where they left off three 
years before. At Quantico the Advance 
Base Force, redesignated the Expedition- 
ary Force in 1921, stood ready to occupy 
and defend an advanced base or to restore 
law and order in a Caribbean republic. 
In that year it included infantry, field 
artillery, signal, engineer, and chemical 
troops, and aircraft. A similar expedi- 
tionary force was planned for San Diego, 
but perennial personnel shortages pre- 
vented the stationing of more than one 
infantry regiment and one aircraft squad- 
ron there during the 1920's.^ 

Nothing seemed changed, but delegates 
of the Great Powers, meeting at Versailles 
to write the peace treaty ending World 
War I, had already taken an action which 
was to have far-reaching consequences for 

* LtCol C. H. Metcalf, A History o/ the United 
States Marine Corps (New York: Putnam's. 
1939), 456-460, 472. 

= CMC AnRepts, 1921-29. 



a future generation of Marines. In the 
general distribution of spoils, the former 
German island possessions in the central 
Pacific had been mandated to the Japa- 
nese. At one stroke the strategic balance 
in the Pacific was shifted radically in 
favor of Japan. That country now pos- 
sessed a deep zone of island outposts. 
Fortified and supported by the Japanese 
fleet, they would constitute a serious ob- 
stacle to the advance of the United States 
Fleet across the Pacific. 

Earl Ellis was one of the first to recog- 
nize the significance of this strategic shift. 
In 1921 he modified his earlier ideas and 
submitted them in the form of Operations 
Plan 712, "Advanced Base Operations in 
Micronesia." In this plan Ellis stressed 
the necessity for seizing by assault the 
bases needed to project the Fleet across 
the Pacific. He envisioned the seizure of 
specific islands in the Marshall, Caroline, 
and Palau groups, some of which were 
actually taken by Marines in World War 
II. He went so far as to designate the 
size and type of units that would be nec- 
essary, the kind of landing craft they 
should use, the best time of day to effect 
the landing, and other details needed to 
insure the success of the plan. Twenty 
years later Marine Corps action was to 
bear the imprint of this thinking : 

To effect [an amphibious landing] in the face 
of enemy resistance requires careful training 
and preparation, to say the least ; and this along 
Marine lines. It is not enough that the troops be 
skilled infantry men or artillery men of high 
morale ; they must be skilled water men and 
jungle men who know it can be done — Marines 
with Marine training.' 

The Commandant, Major General John 
A. Lejeune, and other high ranking 

Marines shared Ellis' views. "The seizure 
and occupation or destruction of enemy 
bases is another important function of the 
expeditionary force," he stated in a lecture 
before the Naval War College in 1923. 
"On both flanks of a fleet crossing the 
Pacific are numerous islands suitable for 
submarine and air bases. All should be 
mopped up as progress is made. . . . The 
maintenance, equipping and training of 
its expeditionary force so that it will be 
in instant readiness to support the Fleet 
in the event of war," he concluded, "I 
deem to be the most important Marine 
Corps duty in time of peace." ^ 

The 1920s, however, were not the most 
favorable years for training in amphibi- 
ous operations. Appropriations for the 
armed services were slim, and the Navy, 
whose cooperation and support was neces- 
sary to carry out landing exercises, was 
more intent on preparing for fleet surface 
actions of the traditional type. Still, a 
limited amount of amphibious training 
was carried out in the first half of the 

During the winter of 1922, a reinforced 
regiment of Marines participated in fleet 
maneuvers with the Atlantic Fleet. Their 
problems included the attack and defense 
of Guantanamo Bay, Cuba, and the island 
of Culebra. In March of the following 
year, a detachment of Marines took part 
in a landing exercise at Panama, and a 
battalion of Marines and sailors practiced 
a landing on Cape Cod that summer. 

Panama and Culebra both witnessed 
landing exercises early in 1924, with a 
Marine regiment participating. This set 
of exercises was the high point of train- 

" OPlan 712, AdvBOps in Micronesia, 1921. 

' MajGen J. A. Lejeune, "The United States 
Marine Corps," MC Oazette, December 1923, 



ing reached in the twenties. It marked 
the advent of serious experimentation 
with adequate landing craft for troops 
and equipment. However, it was most 
notable for the great number of mistakes 
made in the course of the exercises, such 
as inadequate attacking forces, insufficient 
and unsuitable boats, lack of order among 
the landing party, superficial naval bom- 
bardment, and poor judgment in the stow- 
age of supplies and equipment aboard the 
single transport used.* 

The last landing exercise of the era was 
a joint Army-Navy affair held during the 
spring of 1925 in Hawaiian waters. It 
was actually an amphibious command 
post exercise, undertaken at the insistence 
of General Lejeune to prove to skeptical 
Army officers that the Marine Corps could 
plan and execute an amphibious operation 
of greater than brigade size. A force of 
42,000 Marines was simulated, although 
only 1,500 actually participated. It ran 
more smoothly than had the previous 
exercise, but still was handicapped by a 
lack of adequate landing craf t.^ 

Even this meager amphibious training 
came to an end after 1925. New commit- 
ments in Nicaragua, in China, and in the 
United States guarding the mails served 
to disperse the expeditionary forces. By 
1928 the Commandant announced in his 
annual report that barely enough pei"Son- 
nel were on hand at Quantico and San 
Diego to keep those bases in operation." 

^Marines and Amphibious War, 30-32. 

° BriGen Dion Williams, "Blue Marine Corps 
Expeditionary Force," MC Gazette, September 
1925, 76-88 ; LtGen M. B. Twining Itr to ACof S, 
G-3, HQMC, 2oJan57 ; BPlan JA&Nav Exercise, 
192.5, Problem No 3, Blue MarCor ExpedFor, 

'" CMC AnRept, 1928. 

Whatever the shortcomings of the work 
in amphibious doctrine and technique dur- 
ing the 1920's, the Marine Corps scored 
a major triumph when its special interest 
in the field became part of the official 
military policy of the United States. 
Joint Action of the Army and Navy, a 
directive issued by the Joint Board of the 
Army and Navy in 1927, stated that the 
Marine Corps would provide and main- 
tain forces "for land operations in support 
of the fleet for the initial seizure and 
defense of advanced bases and for such 
limited auxiliary land operations as are 
essential to the prosecution of the naval 

Further, in outlining the tasks to be 
performed by the Army and Navy in 
"Landing Attacks Against Shore Objec- 
tives," this document firmly established 
the landing force role of the Marine 
Corps: "Marines organized as landing 
forces perform the same functions as 
above stated for the Army, and because 
of the constant association with naval 
units will be given special training in the 
conduct of landing operations." " 


The recognition of a mission did not 
create the doctrine nor the trained forces 
to carry it out, and, in 1927, neither was at 
hand. In January 1933 the last Marine 
had departed from Nicaragua, and with- 
drawal from Haiti was contemplated. 
Troops were now becoming available for 
training in landing operations, but before 
any real progress could be made, one pre- 
liminary step was essential. A substan- 
tial permanent force of Marines with its 

" The Joint Board, Joint Action of the Army 
and Navy (Washington: GPO, 1927), 3, 12. 



MARINES IN FRANCE IN WORLD WAR I, part of the 4th Marine Brigade of the 2d Infantry 
Division, prepare to move up to the front line trenches. (USMC 4967 ) 

BANDIT-HUNTING PATROL in Nicaragua in 1929 typifies Marine activities between the World 
Wars when the Corps served as a Caribbean riot squad. ( USMC 515283) 



own commander and staff would have to 
be organized for the purpose, otherwise 
training would be constantly interrupted 
by the dispersal of the troops to other 

No one recognized this more clearly 
than the Assistant Commandant, Briga- 
dier General John H. Russell. He as- 
sembled a staff at Quantico to plan the 
organization of a force which could be 
rapidly assembled for service with the 
Fleet. In August of 1933 he proposed to 
the Commandant that the old "Expedi- 
tionary Force" be replaced by a new 
body, to be called either "Fleet Marine 
Force," or "Fleet Base Defense Force." 
The new force, while an integral part of 
the United States Fleet, would be under 
the operational control of the Fleet Com- 
mander when embarked on vessels of the 
Fleet or engaged in fleet exercises afloat 
or ashore. When not so embarked or en- 
gaged it would remain under the Major 
General Commandant. 

Russell's recommendations were 
promptly approved by the Commandant 
and by the Chief of Naval Operations. 
The designation "Fleet Marine Force" 
(FMF) was preferred by the senior naval 
staffs, and the Commandant was requested 
to submit proposed instructions for es- 
tablishing "appropriate command and 
administrative relations between the com- 
mander in Chief and the Commander of 
the Fleet Marine Force." ^^ The decision 
became official with the issuance of Navy 
Department General Order 241, dated 8 
December 1933. 

This directive could well be called the 
Magna Carta of the Fleet Marine Force. 
It stated: 

1. The force of marines maintained by the 
major general commandant in a state of readi- 
ness for operations with the fleet is hereby 
designated as fleet marine force (F. M. F. ), and 
as such shall constitute a part of the organiza- 
tion of the United States Fleet and be included 
in the operating force plan for each fiscal year. 

2. The fleet marine force shall consist of such 
units as may be designated by the major general 
commandant and shall be maintained at such, 
strength as is warranted by the general person- 
nel situation of the Marine Corps. 

3. The fleet marine force shall be available 
to the commander in chief for operations with 
the fleet or for exercises either afloat or ashore 
in connection with fleet problems. The com- 
mander in chief shall make timely recommenda- 
tions to the Chief of Naval Operations regarding 
such service in order that the necessary arrange- 
ments may be made. 

4. The commander in chief shall exercise 
command of the fleet marine force when em- 
barked on board vessels of the fleet or when en- 
gaged in fleet exercises, either afloat or ashore. 
When otherwise engaged, command shall be 
directed by the major generel commandant. 

5. The major general commandant shall de- 
tail the commanding general of the fleet marine 
force and maintain an appropriate staff for 

6. The commanding general, fleet marine 
force, shall report by letter to the commander 
in chief. United States Fleet, for duty in con- 
nection with the employment of the fleet marine 
force. At least once each year, and at such 
times as may be considered desirable by the com- 
mander in chief, the commanding general, fleet 
marine force, with appropriate members of his 
staff, shall be ordered to report to the com- 
mander in chief for conference." 

However significant the creation of the 
FMF may have been in terms of the fu- 
ture, its initial form was modest enough. 
The Commandant was obliged to report 
in August 1934 that the responsibility for 
maintaining ship's detachments and gar- 
risons abroad, and performing essential 

12CN0 Itr to CMC, 12Sep33; Marines and 
Amphibious War, 33-34. 

" Navy Dept GO 241, 8Dec33. 



guard duty at naval shore stations, pre- 
vented the Marine Corps from assigning 
the component units necessary to fulfill 
the mission of the FMF. At this time the 
total number of officers and men in the 
FMF was about 3,000.i* 


With the creation of the FMF the Ma- 
rine Corps had finally acquired the tacti- 
cal structure necessary to carry out the 
primary war mission assigned to it by the 
Joint Board in 1927. The next order of 
business was to train the FMF for the 
execution of its mission. 

But the training could not be very effec- 
tive without a textbook embodying the 
theory and practice of landing operations. 
No such manual existed in 1933. There 
was a general doctrine by the Joint Board 
issued in 1933, and, though it offered 
many sound definitions and suggested 
general solutions to problems, it lacked 
necessary detail. 

In November 1933, all classes at the Ma- 
rine Corps Schools were suspended, and, 
under the guidance of Colonel Ellis B. 
Miller, Assistant Commandant of the 
Schools, both the faculty and students 
set to work to write a manual setting forth 
in detail the doctrines and techniques to 
be followed in both training and actual 
operations. Under the title, Tentative 
Manual for Landing Opera,tions^ it was 
issued in January 1931. 

On 1 August 1934, the title was changed 
to Manual far Naval Overseas Operations 
and some changes were effected in the 
text. A few months later this publica- 
tion, now retitled Tentative Landings Op- 
erations MamMol, was approved by the 

Chief of Naval Operations for "tempo- 
rary use ... as a guide for forces of 
the Navy and the Marine Corps conduct- 
ing a landing against opposition."^' In 
mimeographed form it was given rela- 
tively limited distribution within the 
Navy, but wide distribution within the 
Marine Corps. Comments were invited. 

The doctrine laid down in this remark- 
able document was destined to become the 
foundation of all amphibious thinking in 
the United States armed forces. The 
Navy accepted it as official doctrine in 
1938 under the title of Fleet Training 
Publication 167, and in 1941 the War De- 
partment put the Navy text between 
Army covers and issued it as Field Man- 
ual 31-6. 

Remarkable as it was, the Marine am- 
phibious doctrine was largely theory when 
it was first promulgated at Quantico in 
1934. To put the theory into practice, 
major landing exercises were resumed. 
They were held each winter from 1935 
through 1941 on the islands of Culebra 
and Vieques in conjunction with fleet ex- 
ercises in the Caribbean, or on San Clem- 
ente off the California coast. A final ex- 
ercise of the prewar period on a much 
larger scale than any previously at- 
tempted was held at the newly acquired 
Marine Corps base at New River, North 
Carolina, in the summer of 1941. These 
fleet landing exercises provided the prac- 
tical experience by which details of land- 
ing operations were hammered out. 

In light of its importance, here might be 
as good a place as any to consider briefly 
the more basic aspects of this doctrine as 
conceived in the original manual and mod- 

' CMC AnRept, 1934. 

" NavDept, Tentative Landing Operations 
Manual, 1935, hereinafter cited as Tentative 
Landing Operations Manual. 



ified by experience in fleet exercises up to 
the outbreak of the war. Amphibious op- 
erations and ordinary ground warfare 
share many of the same tactical principles. 
The basic difference between them lies in 
the fact that the amphibious assault is 
launched from the sea, and is supported by 
naval elements. While water-borne the 
landing force is completely powerless and 
is dependent upon the naval elements for 
all its support : gunfire, aviation, transpor- 
tation, and communication. In this ini- 
tial stage only the naval elements have the 
capability of reacting to enemy action. As 
the landing force, however, is projected 
onto the beach, its effectiveness, starting 
from zero at the water's edge, increases 
rapidly until its strength is fully estab- 
lished ashore. 


This basic difference between land and 
amphibious operations created a problem 
in command relationships which has 
plagued amphibious operations from earli- 
est times. During the initial stage when 
only naval elements have the capability 
of reacting to enemy action it has been 
generally and logically agreed that the 
over-all command must be vested in the 
commander of the naval attack force. It 
has, however, not been so generally agreed 
in the past that once the landing force 
is established ashore and capable of exert- 
ing its combat power with primary reliance 
on its own weapons and tactics that the 
landing force commander should be freed 
to conduct the operations ashore as he sees 

The authors of the Tentative Landing 
Operations Manual, writing in 1934, evi- 
dently did not foresee that this particular 
aspect of command relations presented 

a problem that required resolution.^® They 
simply defined the "attack force" as all the 
forces necessary to conduct a landing op- 
eration and added that the attack force 
commander was to be the senior naval of- 
ficer of the fleet units making up the at- 
tack force. His command was to consist 
of the landing force and several naval com- 
ponents, organized as task groups for the 
support of the landing. These included, 
among others, the fire support, transport, 
air, screening, antisubmarine, and recon- 
naissance groups. The commanders of the 
landing force and of the several naval 
task groups operated on the same level 
under the over-all command of the attack 
force commander throughout the opera- 

This initial command concept was des- 
tined to undergo a number of modifications 
and interpretations which will be discussed 
in this history as they occur. The first im- 
portant change did not come about until 
toward the close of the Guadalcanal cam- 


There is nothing new in the concept of 
using the fire of ships' guns to cover an 
amphibious landing of troops during its 
most vulnerable phase : before, during, and 
after the ship-to-shore movement. Our 

" Unless otherwise noted the material in the 
remainder of this chapter is derived from Ten- 
tative Landing Operations Manual; FTP-167, 
Landing Operations Doctrine, U. 8. Navy (Wash- 
ington : OflSee of the CNO, 1938) and changes 1 
& 2 thereto ; 1st MarBrig Itr to CMC, 5Jun39 and 
end (a) thereto; 1st MarBrig Flex 6 Rept, 
"Notes from Critique for Makee Learn Problem 
at Culebra, 14-15Feb40;" 2d MarBrig Minor 
Landing Exercises Rept, San Clemente Island, 
Calif, 17Apr-6May39. 

" See Part VI of this history. 



own history contains many examples of 
this technique, notably: two landings of 
U. S. troops in Canada during the War of 
1812 (York and Niagara Peninsula, sum- 
mer 1813) ; General Scott's landing at 
Vera Cruz in 1847 during the Mexican 
War; several amphibious operations dur- 
ing the Civil War, e. g.^ Fort Fisher in 
1865; and Guantanamo Bay during the 
Spanish-American War in 1898. 

However, the evolution of modern 
weapons posed difficult problems of a 
technical nature, and the much belabored 
Gallipoli operation seemed to indicate 
that these were insoluble. High- 
powered naval guns, with their flat tra- 
jectory and specialized armor-piercing 
ammunition, proved no true substitute for 
land-based field artillery, and much study 
and practice would be required to develop 
techniques which would make them even 
an acceptable substitute. 

Nevertheless, a rudimentary doctrine 
concerning naval gunfire support evolved 
during the years between 1935 and 1941. 
But it evolved slowly and none too 
clearly. Experimentation indicated that 
bombardment ammunition, with its sur- 
face burst, was better suited to fire mis- 
sions against most land targets, while 
armor-piercing shells could be employed 
to good effect against concrete emplace- 
ments and masonry walls. The types of 
ships and guns best adapted to perform 
specific fire missions — close support, deep 
support, ^ counterbattery, interdiction, 
etc. — were determined. And some prog- 
ress was made in fire observation tech- 

Three types of observeis were provided 
for : aerial, shipboard and, once the first 
waves had landed, shore fire control par- 
ties. For the greater part of this period 
the latter were made up of personnel of 

the firing ships, inexperienced in such 
work, untrained, and wholly unfamiliar 
with the tactical maneuvers of the troops 
they were supporting. Not until 1941 
were trained Marine artillery officers with 
Marine radio crews substituted, the naval 
officers then serving in a liaison capacity. 

Other considerations of a naval nature 
served as further limiting factors on the 
NGF support concept. The necessity for 
the support ships to have a large propor- 
tion of armor-piercing projectiles readily 
available with which to fight a surface 
action on short notice restricted the ac- 
cessibility of and limited the amount of 
bombardment shells carried. In turn, the 
probability of enemy air and submarine 
action once the target area became known 
caused much apprehension in naval minds 
and dictated the earliest possible depar- 
ture of the firing ships from the objec- 
tive. An example of this apprehension 
at work came to the fore early in tlie 
Guadalcanal campaign.^* 

Furthermore, tradition dies hard in 
any service. The traditional belief that 
warships exist for the sole purpose of 
fighting other warships dates far back in 
history, with one of its leading exponents 
the great Lord Nelson with his oft-quoted 
dictum : "A ship's a fool to fight a fort." 
This supposed vulnerability of surface 
vessels to shore-based artillery remained 
very much alive in the minds of naval 
planners. So they dictated that support 
ships should deliver their fires at maxi- 
mum range while traveling at high speed 
and maneuvering radically — not exactly 
conducive to pin-point markmanship.^" 

In sum, these considerations,' tiie starting con- 
cept of naval gunfire support with which we 

'^ See Part VI, Chap 2, of this history. 
"Marines and Amphibious War, 38. 



entered World War II, added up to this : a bom- 
bardment of very short duration, delivered by 
ships firing relatively limited ammunition al- 
lowances of types often not well suited to the 
purpose, from long ranges while maneuvering 
at high speeds. Obviously, the best that could 
be expected would be area neutralization of 
enemy defenses during troop debarkation and 
the ship-to-shore movement, followed by a lim- 
ited amount of support on a call basis, with this, 
too, to be withdrawn as soon as field artillery 
could be landed.^" 

Area neutralization — that was the basic 
concept, with deliberate destruction fire 
ruled out. A blood bath would be re- 
quired to expunge this from "The Book." 


As the Marine Corps developed the var- 
ious techniques contributing to a smooth 
landing operation, it had to give more 
and more consideration to the fast 
growth of military aviation as a powerful 

Even the original Tentative Landing 
Operations Manual considered the vulner- 
able concentrations of troops in trans- 
ports, landing boats, and on the beach 
and called for a three-to-one numerical 
superiority over the enemy in the air. 
Later, in FTP-167^ the ratio was increased 
to four-to-one, primarily to wipe the 
enemy air threat out of the skies and 
secondarily to shatter the enemy's beach- 
head defense and to cut off his reinforce- 

Considerable emphasis was placed, how- 
ever, on direct assistance to the troops 
themselves. This included such support- 
ing services as guiding the landing boats 
to the beach, laying smoke screens, and 
providing reconnaissance and spotting for 

™ SM-67, Naval Ounfire in Amphibious Opera- 
tions (Quantico: MCEC, MCS, 195.'5), 2. 

naval gunfire and artillery. Most impor- 
tantly, it included rendering direct fire 
support to the landing force until the 
artillery was ashore and ready to fire. 

For this air war, employment of Ma- 
rine squadrons on carriers was considered 
ideal but, due to a limited number of car- 
riers, was not always a practical possibil- 
ity. Planners even considered moving 
Marine planes ashore in crates and as- 
sembling them, after the ground troops 
had seized an airfield. 

Hence, the Tentative Landing Opera- 
tions Manual called for the Navy to carry 
most of the initial air battle. Marine 
pilots, however, might be employed with 
Navy air units. Actually, in order to exer- 
cise Marine air, most of the early training 
landings had to be scheduled within 
round trip flying distance of friendly air- 
fields. Although by 1940 Marine carrier 
training operations were becoming rou- 
tine, the heavy reliance upon Navy car- 
rier air over Marine landings lasted 
throughout the war. 

As noted before, close coordination of 
air with ground received great emphasis 
in the Marine Corps. Even in Santo 
Domingo and Haiti and later in Nicara- 
gua, Marine pilots reconnoitered, strafed, 
and bombed insurgent positions, dropped 
supplies to patrols, and evacuated 
wounded. The Tentative Landing Oper- 
ations Manual incorporated this team- 
work into its new amphibious doctrine, 
and the landing exercises of the late SO's 
developed aviation fire power as an im- 
portant close ground support weapon. By 
1939, Colonel Roy S. Geiger advocated and 
other Marine Corps leaders conceded that 
one of the greatest potentials of Marine 
aviation lay in this "close air support." 

The challenge became that of applying 
the fire power of Marine air, when needed, 



to destroy a specific enemy front line posi- 
tion without endangering nearby friendly 

Refinement of this skilled technique as 
we know it today was slow because of 
many factors. There was so much for 
pilots to learn about rapidly developing 
military aviation that close air support 
had to take its place in the busy training 
syllabus after such basic drill as aerial 
tactics, air to air gunnery, strafing, bomb- 
ing, navigation, carrier landings, and 
communications, and constant study of the 
latest in engineering, aerodynamics, and 
flight safety. 

Also, whenever newer, faster, and 
higher flying airplanes trickled into the 
Marine Corps in the lean thirties, they 
were found to be less adaptable for close 
coordination with ground troops than the 
slower, open cockpit planes which sup- 
ported the patrol actions of Nicaragua. 

In Nicaragua the aviator in his open 
cockpit could idle his throttle so as to 
locate an enemy machine gun by its sound, 
but in the maneuvers of 1940 pilots flash- 
ing by in their enclosed cockpits found it 
difficult to see what was going on below 
or even to differentiate between friendly 
and "enemy" hills.^^ In Nicaragua, the 
Marine flier was most often an ex-infan- 
tryman, but 10 years later many of the 
new Navy-trained Marine aviators were 
fresh from college and knew little about 
ground tactics. The lack of a real enemy 
to look for, identify, and to shoot at hin- 
dered attempts at precision, especially 
since air-ground radio was not yet as re- 

liable as the old slow but sure system 
where pilots read code messages from 
cloth panels laid on the ground or 
swooped down with weighted lines to 
snatch messages suspended between two 

The main key to development of close 
air support lay in reliable communications 
to permit quick liaison and complete un- 
derstanding between the pilot and the 
front line commander. Part of the solu- 
tion lay in more exercises in air-ground 
coordination with emphasis on standard- 
ized and simplied air-ground communica- 
tions and maps. By 1939 an aviator as an 
air liaison officer was assigned to the 1st 
Marine Brigade Staff. While both artil- 
lery and naval gunfire, however, employed 
forward observers at front line positions, 
air support control was still being chan- 
neled slowly through regimental and bri- 
gade command posts.^^ In the same year 
one squadron sent up an air liaison officer 
in the rear seat of a scouting or bombing 
plane to keep abreast of the ground situa- 
tion and to direct fighter or dive bomber 
pilots onto targets by means of radio.^' 
This was better but not best. 

Meanwhile, war flamed up in Europe. 
Navy and Marine planners took note as 
the Germans drove around the Maginot 
line with their special air-ground "ar- 
mored packets" in which aviation teamed 
up with the fast, mobile ground elements 
to break up resistance." By this time the 
Marines were working on the idea of plac- 

"' From Culebra came the report, ''1st MAG 
as a whole performed in a creditable manner, al- 
though at one stage they were impartial in their 
attacks." 1st MarBrig Flex 6 Rept, ."Notes from 
Critique for Makee Learn Problem at Culebra, 

^ LtGen Julian C. Smith interview by HistBr, 
G-3, HQMC, 25JU156. 

"Col R. D. Moser interview by HistBr, G-3, 
HQMC, 31Aug56. 

" WD G-2 Memo for C/S, 23Sep41, I. B. 130, 
Air-GrdOps, Tab C; CinCLant Flex 6 Rept, 
13Jun40, 14-1.5. 



ing radio-equipped "observers" on the 
front lines to control air support for the 
troops. But the Leathernecks were al- 
ready in the war before the first standard- 
ized Navy-Marine Corps instructions on 
their employment appeared.^^ Also at 
that time, on Guadalcanal certain infan- 
try officers were given additional duty as 
regimental "air forward observers." 
They were coached on the spot by aviators 
of the 1st Marine Aircraft Wing.^® 


The ship-to-shore movement was vis- 
ualized by the Tentative Landing Opera- 
tions Manual in a manner which resem- 
bled closely a conventional attack in land 
warfare: artillery preparation, approach 
march, deployment, and assault by the in- 
fantry. It stressed that this movement 
was no simple ferrying operation but a 
vital and integral part of the attack itself 
and demanded a high order of tactical 
knowledge and skill. 

The two major problems in the ship-to- 
shore movement are the speedy debarka- 
tion of the assaulting troops and their 
equipment into the landing boats and the 
control and guiding of these craft to their 
assigned beaches. To facilitate the first, 
the Tentative Landing Operations Man- 
ual directed that each transport on which 
combat units were embarked should carry 
as a minimum sufficient boats to land a 
reinforced infantry battalion.^^ Thus 

" USN, CSP-1536, 5Sep42. 

"" 1st MarDiv, Final Report on Guadalcanal 
Operation, ljul43. Phase V, Annex D, OPlan 2-42, 
5. The directive on appointing air forward ob- 
servers was dated 20ct42. 

" This general concept that troops and their 
landing craft should be transported together 
to the objective area remained valid through- 

each transport and its accompanying 
troops would be tactically self-sufficient 
for the assault landing, and the loss of one 
ship would not be a crippling blow. To 
expedite their debarkation the Marines 
generally went over the side via cargo nets 
rigged at several stations on the ship. 

To solve the second major problem in 
the ship-to-shore movement, that of con- 
trolling and guiding the landing craft to 
their proper beaches, the Tentative Land- 
ing Operations Manual provided for: (1) 
marking the line of departure with buoys 
or picket boats; (2) a designated control 
vessel to lead each boat group from the 
rendezvous area to the line of departure, 
towing the boats in fog, smoke, or dark- 
ness, if necessary; (3) wave and alternate 
wave guide boats; (4) each boat to carry 
a signboard with its assigned letter and 
number indicating its proper position in 
the formation; and (5) for a guide plane 
to lead the boat waves in. 

The system for the control of the ship- 
to-shore movement was still substan- 
tially the same as prescribed in the Ten- 
tative Landing Operations Manual when 
the Marines made their first amphibious 
landing of World War II at Guadalcanal 
on 7 August 1942. 


"Combat unit loading" of transports is 
the key to amphibious logistics as devel- 
oped by the Marine Corps. This is a prac- 
tical process designed to make supplies 
and equipment immediately available to 
the assault troops in the order needed, dis- 
regarding to a large extent the waste of 
cargo space which results. In contrast 
is commercial loading which is equally 

out the war, although at times it was necessary 
to deviate from it. 



practical in utilizing every cubic foot of 
cargo space available but prevents access 
to much of the cargo until the ship is 

Highest priority items for combat unit 
loading vary somewhat with the nature 
and problems of a particular operation. 
Relative priorities must be worked out 
with minute care. The responsibility for 
handling this was given to a Marine offi- 
cer designated transport quartermaster 
(TQM) aboard each amphibious assault 
ship. He had to know not only the weight 
and dimensions of each item of Marine 
gear carried but had to familiarize himself 
with the characteristics of the particular 
ship to which he was assigned : exact loca- 
tion and dimensions of all holds and stor- 
age spaces in terms of both cubic feet and 
deck space. This familiarity required at 
times accurate remeasurement of holds 
and loading spaces as modifications, not 
shown in the ship's plans, had often been 
made in the ship's internal structure. Ini- 
tially, the Tentative Landing Operations 
Manual directed that the TQM should be 
an officer of the unit embarked, but such 
were the variations in ships that it subse- 
quently proved more feasible to assign a 
Marine officer, thoroughly familiar with 
Marine gear, permanently to a paiticular 
ship with which he would become equally 
familiar through experience. 

Practical experience with combat load- 
ing between 193"5 and 1941 generally con- 
firmed the soundness of the doctrines set 
forth in the Tentative Landing Opera- 
tions Manual. Application of these doc- 
trines in the fleet landing exercises was 
limited, however, by several factors, 
chiefly the lack of suitable transports. 
In addition, an uncertainty at times as to 
ports of embarkation and dates of avail- 

ability of ships sometimes entangled plan- 
ning procedures. As a result, there was 
no ideal approximation of wartime com- 
bat loading. 


One of the most serious problems en- 
countered in early landing exercises was 
congestion on the beaches as men and sup- 
plies piled ashore. To keep such a situa- 
tion reasonably in hand requires a high 
degree of control; control difficult to 
achieve under such circumstances, even 
when the enemy remains only simulated. 
Assault troops must push inland with all 
speed not only to expand the beachhead, 
but also to make room for following units 
and equipment to land and to provide 
space in which personnel assigned strictly 
beach functions can operate. 

To solve this problem the Tentative 
Landing Operations Manual provided for 
a beach party, commanded by a naval 
officer called a beachmaster, and a shore 
party, a special task organization, com- 
manded by an officer of the landing force. 
The beach party was assigned primarily 
naval functions, e. g.^ reconnaissance and 
marking of beaches, marking of hazards 
to navigation, control of boats, evacuation 
of casualties, and, in addition, the unload- 
ing of material of the landing force from 
the boats. The shore party was assigned 
such functions as control of stragglers 
and prisoners, selecting and marking of 
routes inland, movement of supplies and 
equipment off the beaches, and assignment 
of storage and bivouac areas in the vicin- 
ity of the beach. The composition and 
strength of the shore party were not set 
forth except for a statement that it would 
contain detachments from some or all of 
the following landing force units: medi- 
cal, supply, working details, engineers, 



military police, communications, and 
chemical. The beach party and the shore 
party were independent of each other, but 
the Tentative Landing Operations Man- 
ual enjoined that the fullest cooperation 
be observed between the beachmaster and 
the shore party commander, and the per- 
sonnel of their respective parties. 

It was not indicated from what source 
"working details" for the shore party 
would come, but in practice, since there 
was no other source, the policy of as- 
signing units in reserve the responsibility 
for furnishing the labor details quickly 
developed. This in effect, however, tem- 
porarily deprived the commander of his 

No realistic test of the shore and beach 
party doctrine took place during the early 
fleet landing exercises. Although some 
material was landed on the beach, it gen- 
erally consisted of rations and small quan- 
tities of ammunition and gasoline. Not 
until 1941 were adequate supplies avail- 
able and the maneuvers on a large enough 
scale to provide a test of logistic proce- 
dures. The results were not encouraging. 
"In January of 1941 . . . the shore party 
for a brigade size landing . . . consisted 
of one elderly major and two small piles 
of ammunition boxes," wrote a Marine of- 
ficer who "suffered" through those years. 
"The ship-to-shore movement of fuel was 
a nightmare. We had no force level 
transportation, [no! engineers and no sup- 
porting maintenance capability worthy of 
the name. In short, the combination of 
the parsimonious years and our own 
apathy had left us next to helpless where 
logistics were concerned." ^® 

Major General H. M. Smith, the land- 
ing force commander at the New River 
exercise in the summer of 1941, reported 
that "considerable delay in the debarka- 
tion of troops and supplies was caused by 
lack of personnel in the Shore and Beach 
Parties .... Roughly, the supplies ex- 
cept for subsistence it was possible to 
land . . . were insufficient to sustain the 
forces engaged for more than three 
days." 3° 

General Smith, who had a deep respect 
for logistics, was determined to correct 
these deficiencies. "It is evident," he re- 
ported to Rear Admiral Ernest J. King, 
Commander in Chief, Atlantic Fleet, 
"that special service troops (labor) must 
be provided for these duties in order to 
prevent reduction of the fighting strength 
of battalion combat teams .... The 
present doctrine results in divided author- 
ity between shore party commanders." 
He i-ecommended that "the beach and 
shore party commanders be consolidated 
into one unit, a Shore Party, under con- 
trol of the landing force." ^^ 

Solution to the problem of divided au- 
thority came from a joint board of Army, 
Navy, Marine Corps, and Coast Guard 
officers appointed by Admiral King. Its 
recommendations closely followed those 
of General Smith and were accepted in 
toto and published on 1 August 1942 as 
Change 2 to FTP 167. The principal 
changes were: (1) joining together of the 
beach and shore parties under the title 
Shore Party, as a component of the land- 
ing force; (2) designating the beach party 
commander as the assistant to the shore 
party commander and his advisor on 

"BriGen V. H. Krulak Itr to ACofS, G-3, 
HQMC, 5Mar57. 

"• CG LantPhibFor PrelimRept to CinCLant on 
New River Exercise 4-12Aug41, 27Aug41. 
" Ibid. 



naval matters; and (3) transferring the 
responsibility for unloading boats at the 
beach from the naval element to the land- 
ing force element of the shore party .^^ 

Marine Corps Headquarters solved the 
labor force problem by adding a pioneer 
(shore party) battalion of 34 officers and 
669 enlisted men to the marine division.'^ 
This change occurred on 10 January 1942, 
too late for the personnel concerned to 
gain practical experience in large-scale 
exercises in the techniques of handling 
vast quantities of supplies or to test the 
adequacy of the strength and organization 
provided. At Guadalcanal this lack came 
close to having serious consequences.^* 

General Smith was not content merely 
to submit his shore party recommenda- 
tions to Admiral King. At his direction, 
the logistics staff of the Amphibious 

• Ibid. 

' Marine Corps T/0 D-94, 10Jan42. 

' See Part VI of this history. 

Force Atlantic Fleet prepared a detailed 
Standing Operating Procedure (SOP) 
covering all phases of logistics. Issued as 
Force General Order No. 7-42, SOP for 
Supply and Evacuation, it served as the 
basic guide to combat loading and shore 
party operations during the Guadalcanal 

By 7 December 1941 the Marine Corps 
had made long strides towards amphib- 
ious preparedness. It had a doctrine 
which had been tested in maneuvers and 
found to be basically sound. Many of 
the errors in implementation had been 
recognized and corrected ; still others were 
awaiting remedial action when war broke 
out. But the simulated conditions of the 
maneuver ground were now to be aban- 
doned. The Marines and their doctrine 
were now to submit to the ultimate test 
of war. 

'Krulak, op. cit.; Twining, op. cit. 


Development of Landing Craft 


The amphibious warfare doctrine la- 
boriously developed by Marines between 
the two World Wars could never have 
been successfully executed without special 
equipment to transport the assaulting 
troops and their supplies from ship to 
shore and to land them on an enemy- 
defended beach. 

No one was more aware of the need 
for such equipment than the Marines. 
Shortly after the end of World War I 
they induced the Navy to undertake de- 
sign studies on two landing craft, one for 
personnel and one for materiel. Troop 
Barge A, as the first of these types was 
called, was tried out at Culebra in the 
winter of 1923-24. A shallow draft, 
twin-engined, 50-foot craft with a rated 
speed of about 12 knots and a carrying 
capacity of 110 fully equipped Marines, 
it had good beaching qualities and could 
retract from the beach with aid of a stern 
anchor. Three years later the second 
type, a 45-foot artillery lighter, was built 
and tested. Equipped with two parallel 
hinged ramps in the stern, it could be 
beached successfully stern-to and 155mm 
guns and other pieces of heavy Marine 
equipment unloaded. It lacked a power 
plant, however, and had to be towed by 
another craft.^ 

'LtGen K. E. Rockey Itr to ACofS, G-3, 
21Jun57; 2dLt W. B. Trundle rept on experi- 
ments with Beetle Boat to CG, INIarCorExpedFor, 
3Mar24; Senior Member, BoatCom Itr to Pres, 

Another item of equipment tried out in 
1924 was the Christie "amphibian tank." 
Afloat, this unusual machine was driven 
by twin-screw pi'opellers at a rated speed 
of seven knots. On short, as a tractor, it 
could make 15 mph ; or, where good roads 
were available, the demountable tracks 
could be removed, and on wheels it could 
do 35 mph. It functioned well enough on 
land and in the sheltered waters of rivers. 
But in the open sea, under conditions that 
must be realistically anticipated for an as- 
sault landing, it proved so unseaworthy 
that the Marine Corps directed its atten- 
tion to other types. 

The construction of these types of 
amphibious equipment constituted a be- 
ginning, however humble, towards the 
solution of the problem of transporting 
troops and equipment from ship to shore. 
But a shortage of funds made it impossible 
to follow up these developments until 
1935, when appropriations became more 
plentiful as a result of the naval expan- 
sion program begun in the first Roosevelt 


With the publication of the Tentative 
Landing Operations Manual in 1934 and 
the resumption of landing exercises the 
following year, work on the landing craft 

MCEB, 21Jul36, both in War Plans See HQMC 
files, folder "Landing Boats and Barges, 1924- 
1939," hereinafter cited as War Plans Files, 




was resumed. Three types of boats for 
landing operations were contemplated by 
Marine planners of the mid-thirties. 
These included fast, small, surf boats to 
lift the leading waves; standard Navy 
boats and life boats of merchant vessels 
for the bulk of troops; and barges and 
lighters for heavy material. - 

Steps to solve the first problem, pro- 
vision of special troop landing boats, were 
initiated in 1935. The Marine and Navy 
officers who tackled the problem that year 
had to start pretty much from scratch, for 
Troop Barge A, a promising early develop- 
ment, fell victim to the size and weight re- 
strictions imposed by naval ships in those 
days. Navy thinking and planning for 
the development of amphibious equipment 
was restricted by the types of ships then 
serving the fleet. Troop transports were 
practically nonexistent, so it was planned 
as an emergency measure to lift Marine 
landing forces in battleships and cruisers. 
A length of 30 feet, the size of davits on 
these ships, and a weight of five tons which 
was the maximum capacity of the davits, 
were therefore imposed as basic require- 
ments for all new landing craft. 

In an effort to explore the suitability of 
existing commercial craft for landing op- 
erations, the Navy, at the request of the 
Marine Corps, agreed to test as wide a 
variety of small craft from the yards of 
private builders as the limited funds avail- 
able would permit. Bids were advertised, 
and nine replies were received, four of 
which met with the approval of the Ma- 
rine Corps Equipment Board and were 

' CMC to Chief BuC&R, 24Nov36, 2d endorse- 
ment to ComlnCh Itr to CNO, 140ct36. War Plans 
Files, 1924-39. 


Tests of these approved types were con- 
ducted at Cape May, New Jersey, in the 
summer of 1936. But the experiments fell 
short of the original intention, "to test as 
wide a variety of forms as was practicable,'' 
because Andrew Higgins, a New Orleans 
boat builder with a promising design, de- 
clined to submit a bid. In 1926 Higgins 
had designed a special shallow draft craft 
called the Eureka for the use of trappers 
and oil drillers along the lower Mississippi 
and Gulf coast. It had a tunnel stern to 
protect the propeller and a special type 
of bow, called by Higgins a "spoonbill," 
which enabled it to run well up on low 
banks and beaches and retract easily. In 
1934 the inventor had visited Quantico to 
interest Marines in his boat, and the Navy 
was now particularly anxious to test it 
with other comparable types of small 

The four boats which showed up at Cape 
May for the test were of two general types. 
The sea skiff, a boat employed by Atlantic 
coast fishermen, was represented by the 
Bay Head, Red Bank, and Freeport boats. 
This type appeared in theory to offer a so- 
lution to the landing craft problem, as it 
was normally launched and landed 
through the heavy surf of the Atlantic 
beaches in fishery work. The other boat, 
a sea sled built by the Greenport Basin and 
Construction Company, was a high speed 
craft not normally employed in surf nor 
landed on beaches. The test board, com- 
prising representatives of the Navy gen- 
eral line. Bureau of Construction and Re- 
pair, Bureau of Engineering, the Coast 
Guard, and the Marine Corps, reported 
that none of the boats were wholly satis- 
factory. They eliminated the sea sled en- 

' Asst Chief BuC&R Itr to Higgins Industries, 
210ct36, S82-3 (15) BuShips files. 



EXPERIMENTAL AMPHIBIAN TRACTOR developed for the Marine Corps in 1924 began the long 
line of test vehicles that culminated in the LVT. (USMC 13562 ) 

AN EARLY VERSION OF THE LANDING CRAFT used m World War II which resulted from 
joint Navy-Marine Corps experiments in the 1920's and 30' s. ( USMC 515227 ) 

tirely and recommended that the three re- 
maining craft be modified and sent to the 
Fleet for further tests.^ 

' CMC to Chief BuC&R, 24Nov36, 2d endorse- 
ment to Cominch Itr to CNO, 140ct36, War Plans 
Files, 1924-39. 

These tests took place at Culebra during 
Flex 4 in the winter of 1938. Though 
superior in speed and beaching ability to 
standard Navy boats, the modified fishing 
craft still had serious drawbacks. Owing 
to their exposed rudders and propellers 



they tended to dig in when retracting. 
They were so high forward that Marines 
debarking had to drop 10 feet from the 
bow to the beach. They were, moreover, 
all unsuitable for lowering and hoisting.® 

In the light of the drawbacks revealed 
by tests, the Bureau of Construction and 
Repair undertook the construction of a 
boat embodying all the best features of the 
fishing craf t.^ This was the beginning of 
a long and unsuccessful effort by the Bu- 
reau to develop a satisfactory landing 
craft. The "Bureau Boat" in various 
forms showed up regularly at Fleet Land- 
ing Exercises from 1939 through 1941, but 
efforts to get the "bugs" out of its design 
were abandoned in 1940. 

Experiments with standard Navy ships' 
boats proceeded simultaneously with the 
development of special types. From the 
first they proved unsatisfactory. After 
five of them foundered in a four-foot surf 
at San Clemente during Flex 3, efforts 
to adapt standard Navy boats for beach 
landings were abandoned. The fact was 
that, having been designed for other pur- 
poses, none of them were suitable for 
beaching operations. As the Command- 
ing Officer of the 5th Marines concluded : 
"Navy standard boats are totally unsuited 
for landing troops of the leading waves, 
even under moderate surf conditions. 
They are in no sense tactical vehicles, 
lacking in speed and maneuverability and 
are extremely difficult to handle in surf. 

" CG 1st MarBrig Flex 4 Rept, 12Mar38 ; BriGen 
V. H. Krulak Itr to Head HistBr, G-3, HQMC, 
lFeb57, w/attached comments. 

'CMC to Chief BuC&R, 24Nov36, 2d endorse- 
ment to Cominch Itr to CNO, 140ct36, War Plans 
Files, 1924-39. 

They do not permit the rapid debarkation 
of troops at the water's edge." * 

By 1938 a beginning had been made 
towards the solution of the landing craft 
problem. As a result of the early experi- 
ments the Marines had proved to their 
own satisfaction what they had suspected 
all along — that none of the standard Navy 
boats could be adapted satisfactorily for 
the landing through surf of troops or 
heavy equipment. Nor were the experi- 
mental models based on commercial craft, 
though superior to Navy boats, a satis- 
factory means for landing of assault 
waves on a defended beach. These results, 
though negative in character, at least 
cleared the way for concentrating develop- 
ment on specially designed landing craft. 

The fruitful line of development came 
into view with the re-entrance of Andrew 
Higgins into the picture. In October 
1936, about a year after declining to bid 
on the experimental landing boat contract, 
Higgins had written the Navy offering 
his Eureka as a troop landing craft. As 
funds for the purchase of experimental 
boats had been exhausted, the Navy was 
unable to purchase the Higgins craft at 
that time.^ 

A year later Commander Ralph S. 
McDowell, Avho was responsible for land- 
ing craft development in the Bureau of 
Construction and Repair, learned of the 
Eureka boat. He wrote Higgins inviting 
him to visit the Navy Department and 
discuss this boat if he ever came to Wash- 
ington. Higgins and his naval architect 

' CO 5th Mar Flex 3 Rept, 26Feb37. 
" Asst Chief BuC«&R Itr to Higgins Industries, 
210ct36, S82-3 (15) BuShips files. 



caught the first train for Washington. 
They spent about a week in McDowell's 
office working out a conversion of the 
standard Eurelca into a landing craft. As 
funds for the purchase of experimental 
boats had been exhausted, the Navy De- 
partment at first refused to purchase the 
Higgins craft. But after the inventor 
offered to build a boat for less than cost, 
the Department relented, found the neces- 
sary funds, and gave Higgins a contract 
for one boat. Higgins delivered it to Nor- 
folk in 30 days.'" 

The Eureka was tested in surf at 
Hampton Eoads in the spring of 1938 " 
and made its first maneuver appearance at 
Flex 5 in 1939 where it competed against 
several Bureau boats and the by now 
venerable fishing craft. Marines were en- 
thusiastic about its performance. "The 
Higgins boat gave the best performance 
under all conditions. It has more speed, 
more maneuverability, handles easier, and 
lands troops higher on the beach," re- 
ported the commanding officer of the 1st 
Battalion, 5th Marines. "It also has 
greater power in backing off the beach; 
not once was the boat observed having 
difficulty in retracting."' ^^ 

Lieutenant Commander R. B. Daggett, 
the representative of the Bureau of Con- 
struction and Repair at Flex 5, did not 

" Capt Ralph S. McDowell, USN, interview by 
HistBr, G-3, HQMC, 19Jun.")7 ; Asst Chief 
BuC&R Itr to Higgins Industries, 210ct36, S82-3 
(15) Bn Ships files. 

" LCdr G. H. Bahm Itr to CNO, 7Jnn38, S82-3 
(15) BuShips files. 

" CO 1/5 Flex 5 Kept No 14 to CG 1st MarBrig, 

share the Marines* enthusiasm for the 
Higgin's EureJi-a. "The Higgins . . . boat 
is too heavy. . . . The speed is too slow. 
. . . All the Higgins boats have 250 horse- 
power with accompanying excessive gaso- 
line consumption for the speed ob- 
tained," " he reported to his bureau. 

Daggett's preference was for a modified 
Bureau boat built by the Welin Company. 
The other Bureau types and the fishing- 
boats he found unsatisfactory, and as the 
Marine Corps and the Bureau were in 
agreement, on this point at least, these 
craft were discarded. 

Neither the Marine Corps nor the Bu- 
reau of Construction and Repair was to 
have tlie last word at Flex 5. The Com- 
mander Atlantic Squadron, as represented 
by his Landing Boat Development Board, 
recommended further tests for the Bureau 
and Eureka craft. Accordingly at Flex 6 
the following year the drama was re- 
enacted. Again the Marines declared the 
Eureka to be "the best so far designed." 
The Atlantic Squadron, shifting slightly 
from dead center, decided that the Hig- 
gins "was the best all-around boat for the 
purpose intended . . . [but] the Bureau 
was almost as good." " 

By 1940 money for naval purposes was 
beginning to be more plentiful, and the 
Navy was now willing to purchase landing 

" LCdr R. B. Daggett memo to Chief BuC&R, 
13Feb39, end to Chief BuC&R Itr to CNO, 
16Feb39, War Plans Files, 1924-39. 

1* Comments & Recommendations of Umpires 
and Observers, Flex 6, January-March 1940 ; Ex- 
perimental Landing Boat Group OflBcer Itr to 
ComLantRon, 10Mar40, War Plans Files, 



craft in quantity. But in view of the 
fact that the Fleet was unable to make a 
clear-cut recommendation for either the 
Bureau or Higgins types, the Navy let 
contracts for the first 64 landing craft on 
a fifty-fifty basis." 

The question was finally settled in Sep- 
tember 1940. The Navy was now con- 
verting large merchant ships for use as 
troop transports. These ships were 
equipped with davits capable of handling 
36-foot boats, and as the Eureka of 36- 
foot length had twice the capacity of the 
30-footer then in service and could make 
the same speed without an increase in 
horsepower, the Navy decided to adopt 
the larger as standard.^*' 

After five years of work the Marines 
finally had the landing craft they wanted. 
The one featui-e that kept the Higgins 
boat from fulfilling the ideal that they 
had built uj) in their minds was the diffi- 
culty of emptying it on the beach : all 
troops, equipment, and supplies had to be 
unloaded over the fairly high sides. Dur- 
ing a visit to Quantico in April 1941, 
Higgins was shown a picture of a Japa- 
nese landing craft with a ramp in the bow 
by Major Ernest E. Linsert. Higgins be- 
came enthusiastic about the idea and re- 
turned to New Orleans determined to ex- 
amine the possibility of installing a ramp 
in the bow of his 36-foot Eureka. Lin- 
sert, who was serving as Secretary, Ma- 
rine Corps Equipment Board, recom- 
mended to the President of the Board, 
Brigadiet General Emile P. Moses, that 

the Marine Corps procure a ramp-bow 
36-foot Eureka. Upon receiving the ap- 
proval of Marine Corps Headquarters, 
Moses and Linsert went to New Orleans 
to assist Higgins, who had agreed to make 
a prototype, converting a standard 36- 
foot Eureka into a ramp-bow boat at his 
own expense. 

On 21 May, informal tests were con- 
ducted on Lake Pontchartrain. The new- 
craft proved to be seaworthy. She 
beached and retracted with ease, and 
while on the beach the ramp was lowered 
and personnel and a light truck were de- 
barked and reembarked. On the recom- 
mendation of the Navy Department Con- 
tinuing Board for the Development of 
Landing Boats," a special board of Ma- 
rine Corps and Bureau of Ships officers 
was appointed to conduct official accep- 
tance tests. With General Moses as sen- 
ior member the board carried out the tests 
during the first week in June. The ramp- 
bow craft passed with flying colors." 

Thus was born the precursor of the 
LCVP (landing craft vehicle, personnel), 
the craft which, in the opinion of Gen- 
eral H. M. Smith, '' . . . did more to win 
the war in the Pacific than any other 
single piece of equipment.'' " 

^ DeptContBd for Dev of Landing Boats Rept 
to CNO, 18May40, and BuShips Itr to Cdt 5th 
Naval Dist, 8Jul40, both C-S82-3(15) BuShips 

''CNO Itr to Chief BuShips, 23Sep40, 2455- 
130-60 HQMC files. 

" This board had been created by SecNav on 
12Jan37 to coordinate landing craft develop- 
ment. It was composed of representatives of 
the CNO, BuC&R, BuEng, and MarCorps. 

"LtCol E. E. Linsert interview by HistBr, 
HQMC, 3Jun57, hereinafter cited as Linsert in- 
terview; BriGen E. P. Moses msg to CMC, w/ 
endorsements, 21May41, 2455-130-60 HQMC 
files; MajGen E. P. Moses Itr to ACofS, G-3, 
HQMC, llApr57 ; CNO Itr to CMC et a1, 2Jun41, 
2455-130-60 HQMC files. 

'"Gen H. M. Smith, Coral and Brass (New 
York: Charles Scribner's Sons, 1949), 72. 




The design of a successful tank lighter 
proved a longer and more difficult process 
than did the development of the person- 
nel landing craft. The old 45- foot ar- 
tillery lighter, developed in 1927, was 
considered to have a limited usefulness 
for landing heavy equipment in the later 
stages of an operation, but the Marine 
Corps hoped to obtain a lighter, self-pro- 
pelled craft particularly suited to landing 
tanks during the early stages.™ 

As a stop-gap measure. Marines at 
Quantico came up with a device to adapt 
the standard Navy 50-foot motor launch 
for landing light vehicles and artillery. 
"Boat Rig A," this contraption was called. 
It consisted of a platform fitted within 
the hull of the boat, together with a port- 
able ramp by means of Which the vehicle 
could go ashore over the bow when the 
craft beached. The ramp was carried 
into the beach broken down, where it was 
assembled and hitched up for debarkation. 
This completed, it would be disengaged 
and left on the beach to accommodate the 
next boat coming in. The ramp could be 
assembled and made ready for use by 
eight men in about 10 minutes. On sub- 
sequent trips, it took about four minutes 
to connect the ramp to the boat. Under 
ideal conditions vehicles up to five tons 
in weight could be landed from a 50-foot 
motor launch using Boat Rig A. In calm 
water Boat Rig A worked fairly well, but 
when it was tried out at Culebra in 1935, 
it proved so top heavy that it nearly cap- 
sized in a moderate swell. The experi- 
ment was accordingly written off." 

With the failure of Boat Rig A, the 
Marine Corps turned its attention to de- 
veloping a self-propelled lighter designed 
specifically for landing tanks and heavy 
equipment through the sut-f. In Decem- 
ber 1935 the Commandant requested the 
Bureau of Construction and Repair to 
design such a craft. It was to be capable 
of landing the 9,500-pound Marmon- 
Herrington tank which the Marine Corps 
was then considering. Negotiations 
dragged on for more than a year, until 
in April both the Marine Corps and the 
Bureau had agreed upon a design. A 
38-foot craft, it made its first appearance 
at a fleet landing exercise in 1938.^^ 
The Marines reported it to be "a distinct 
improvement over previous experimental 
designs. It is self-propelled, has suffi- 
cient speed, and is sound and practicable 
in construction. It is equally adaptable 
for landing artillery and is an efficient 
cargo carrier." ^^ 

A 40-footer, built at the Norfolk Navy 
Yard in the autumn of 1938, showed up at 
Culebra the following winter for Flex 5. 
It was used successfully in transporting 
ashore tanks and trucks of the types then 
standard in the Marine Corps. Under the 
conditions encountered at Culebra in 1939, 
both the 38- and 40-foot lighters were 
judged to be ". . . good sea boats, handle 
well, have sufficient power and speed, and 
are capable of retracting themselves from 
the beach by use of their stern anchors. . . . 
Both types . . . proved suitable for landing 
tanks and motor vehicles. The new 

'" CMC to Chief BuC&R, 24Nov36, 2d endorse- 
ment to Cominch Itr to CNO, 140ct36, War 
Plans Files, 1924-39. 

''■ Marines and Amphibious War, 47 ; Tenta- 
tive Landing Operations Manual, 78-80. 

"CMC Itr to Chief BuC&R, 19Dec35, S82- 
3(16) BuShips flies; Senior Member, BoatCom 
Itr to Pres, MCEB, 21Jul36 ; CNO Itr to Chief 
BuC&R, 8Jul36; CMC Itr to Chief BuC&R, 
17Apr37 ; Chief BuC&R Itr to Cdr R. H. English, 
7Apr37, all in War Plans Files, 1934-39. 

'" CG 1st MarBrig Flex 4 Rept, 12Mar38. 



lighter proved superior to the old in re- 
spect to ease and safety of loading in a 
seaway as well as cargo-carrying ca- 
pacity." 2* 

All tank lighter experiments conducted 
up to the end of Flex 5 had been built 
around the Marmon-Herrington tank. 
This vehicle, adopted by the Marine Corps 
in 1935, had been designed to fit within 
the weight limitations imposed by the 
Navy for amphibious equipment. Light- 
ness was just about the only virtue pos- 
sessed by this tank. By 1939 the Marine 
Corps had given up on it and was testing 
the Army light tank for its suitability in 
amphibious operations. As the Army 
tank weighed about 15 tons, it could not 
be carried in any of the tank lighters then 
in existence. The Navy accordingly pro- 
duced a new model 45-feet in length, capa- 
ble of carrying one Army and two Mar- 
mon-Herrington tanks.^^ 

One of the new 45-f ooters was completed 
in time for a trial at Culebra during the 
winter of 1940 in Flex 6. The tests lacked 
somewhat in realism, however, because 
none of the Army-type tanks were avail- 
able. The new lighter performed ade- 
quately as a carrier of the Marmon- 
Herrington tank, for other vehicles, and 
miscellaneous heavy equipment. At the 
end of Flex 6, General Smith recom- 
mended to the Commandant that ". . . . 
20 of the 45-foot lighters be constructed, 
at the earliest practicable date, for use by 
the Atlantic Squadron in landing opera- 
tions." ^^ 

In the fall of 1940 the Navy contracted 
for the construction of 96 45 -foot tank 
lighters. After the contract had been 
awarded, doubt arose as to the seaworthi- 
ness of the basic design. During a land- 
ing exercise in the Caribbean, one of the 
45-footers capsized and sank when the 
Army-type tank it was carrying shifted 
to one side in a moderate sea.^^ 

In the spring of 1941 the Marine Corps 
found itself in urgent need of all the 
lighters it could lay its hands on for use 
in a proposed amphibious landing in the 
Azores.^* None of the 96 lighters ordered 
by the Navy had been delivered, and not 
more than eight or ten were expected in 
time for the operation. Therefore, on 27 
May 1941 the Navy Department Con- 
tinuing Board for the Development of 
Landing Boats recommended that Hig- 
gins be given an opportunity to convert 
one of his 45 -foot Eureka boats into a 
tank lighter by installing a ramp in the 
bow. If this craft met service tests he 
would be awarded a contract for 50 tank 
lighters. The Secretary of the Navy gave 
his approval on 29 May, and Higgins re- 
ceived this order by telephone the next 

Higgins rushed through the conversion, 
completing it in time for testing and ac- 
ceptance during the first week in June by 
the same board of Marine Corps and Bu- 

" CO 1/5 Flex 5 Rept No 14 to CG 1st Mar- 
Brig, 15Mar39. 

"^ BuC&R Itr to Cdt Norfolk Navy Yard, 6Jul39, 
War Plans Files, 19J,0-U. 

'°CG 1st MarBrig Itr to CMC, 29Apr40, War 
Plans Files 19iO-41. 

"ConiLantRon Itr to CNO, 13Dec40, 2455- 
130-60 HQMC files ; Senate Report No. 10, Part 
16, Additional Report of the Special Committee 
Investigating the National Defense Program, 
78th Congress, 2d Session, hereinafter cited as 
Senate 10. 

'* See Part I, Chap 5 of this history. 

^^''CNO Itr to CMC et al, 2Jun41, 2455-130-60 
HQMC files ; Senate 10, 139; Capt R. B. Daggett, 
USN, interview by HistBr, G-3, HQMC, 2aMay57, 
hereinafter cited as Daggett Interview. 



reau of Ships officers who had come to 
New Orleans to test the 36-foot ramp-bow 
Eureka. At the New River exercises that 
summer the Higgins tank lighters proved 
to be of excellent basic design. "They 
were found to be fast, subject to ready 
control and retraction, relatively light, 
and equipped with a reliable power 
plant," reported General Smith.^" They 
also proved to be too hastily constructed. 
The ramps were so weak that several col- 
lapsed, and the sill was too high for effi- 
cient handling of vehicles. Higgins, who 
was present, was confident that he could 
correct the deficiences.^^ 

Before the reports of the New River 
exercises had been received by the Navy 
Department, a contract had been let for 
131 additional tank lighters. These were 
of a 47-foot Bureau design, a prototype 
of which had never been built. As a re- 
sult of the good showing of the Higgins 
tank lighter at New River, this contract 
Avas later reduced to ten. Higgins was the 
low bidder, and built one craft to Bureau 
specifications, although he was convinced 
that the design was unseaworthy. His 
fears proved to be well founded when the 
tests were carried out. By this time, how- 
ever, the tank lighter program had again 
changed direction.^^ 

On 4 October 1941, the Auxiliary Ves- 
sels Board of the Navy had reported that 
there was no lighter capable of landing 
the newly developed Army 30-ton medium 
tank. The Secretary of the Navy directed 

'"CG PhibLant Itr to CinCLant, 9Sep41, 
FMFLant files. 

"'CNO Itr to CMC et al, 2 Jun41, . 2455-130-60 
HQMC files; Daggett Intervieiv; Linsert Inter- 
view; CinCLant Itr to CNO, 70ct41, FMFLant 

^Senate 10, 139-140. 

the Bureau of Ships to remedy this defi- 
ciency. Accordingly, in December exist- 
ing tank lighter contracts were changed to 
provide 50- footers in lieu of the 45-foot 
Higgins and 47-foot Bureau types still to 
be built. Both Higgins and the Bureau 
produced designs of 50-foot craft. Before 
any deliveries could be made. President 
Roosevelt, at a White House Conference 
on 4 April 1942, directed the procurement 
of 600 additional 50-foot tank lighters by 
1 September for the North African op- 
eration. The Bureau of Ships, to meet 
this commitment, ordered 1,100 of its own 

Since this order was earmarked for serv- 
ice in a projected Army operation, the 
Army showed keen interest in a test of the 
two types held near Norfolk on 25 April 
1942. Each carried a 30-ton tank, elabo- 
rately lashed down in the Bureau lighter, 
merely blocked in place in the Higgins. 
Wind velocity ran 18 to 23 miles per hour, 
with wave heights estimated between li/^ 
and 2 feet. Both lighters showed a speed 
of 10 miles an hour over a measured li/^- 
mile course. What happened after that is 
described by the Army observer who made 
the trip in the Higgins type : 

As we neared the [antisubmarine] net it be- 
came apparent that the Navy Bureau-type tank 
lighter was in trouble. She appeared to have a 
tendency to dive when headed into the seas 
and was taking considerable water aboard. She 
stopped several times and members of the crew 
could be seen manning hand pumps and attempt- 
ing to better secure the tank in the lighter. Once 
when under way and making a wide turn, it ap- 
peared that the lighter was going to overturn. 
Some of the crew was seen straddling the higher 
bulwark and the coxswain had left the pilot 
house and was steering the vessel from the rail. 

While this was going on, our [Higgins] lighter 
was standing by, as was a picket boat and two 

Ibid., 157. 



Higgins 36-foot boats. None of these vessels was 
experiencing any difficulty. The Higgins tank 
lighter was maneuvering around in sharp turns 
into the sea, through the wave troughs. 

We then [after Bureau lighter turned back] 
opened the engines up to 1,900 r. p. m. and pro- 
ceeded past Little Creek to Fort Storey. The 
lighter took no water except a little spray. Per- 
formance was excellent in all respects. The 
lighter was beached in the surf and the tank ran 
off onto the beach despite poor handling by the 
coxswain who finally allowed the lighter to 
broach to. In spite of this the vessel had such 
power and retraction qualities [as] to get back 
into deep water. 

As far as comparison of characteristics of the 
types of tank lighters are concerned, it may be 
stated that in the May 25 tests there was no 
comx>arison. . . ." 

As a result of these tests, the Bureau 
hastily notified all yards to shift to the 
Higgins type. Thus the Higgins 50-f ooter 
became the standard tank lighter of the 
Navy, the prototype of the LCM (landing 
craft, mechanized) as the Marines knew it 
in World War II, and as they know it to- 
day in enlarged form. 


Another vehicle which was to play a 
vital role in the amphibious operations of 
World War II was the amphibian tractor 
(amtrack, LVT) . It was built in 1935 by 
Donald Roebling, a wealthy young in- 
ventor living in Clearwater, Florida. The 
"Alligator," as Roebling called his cre- 
ation, was a track-laying vehicle which 
derived its propulsion afloat from flanges 
fixed to the tracks, essentially the prin- 
ciple of early paddle-wheel steamships. 
Originally intended as a vehicle of mercy, 
for rescue work in the Everglades, the "Al- 
ligator" was destined for fame as an in- 
strument of war. 

The Marine Corps first took notice of 
the "Alligator" in 1937, when Rear Ad- 
miral Edward C. Kalbfus, Commander, 
Battleships, Battle Force, U. S. Fleet, 
showed Major General Louis McCarthy 
Little, then commanding the Fleet Marine 
Force, a picture of the strange vehicle ap- 
pearing in Life magazine. General Little 
was quick to grasp its potentialities and 
sent the picture and accompanying article 
to the Conunandant. He, in turn, passed 
it along to the Equipment Board at 

The Marine Corps had not forgotten the 
old Christie amphibian, of such bright 
promise and disappointing performance. 
Here appeared to be a possible answer. 
The Board dispatched its secretary, then 
Major John Kaluf, to Florida to see the 
vehicle perform and to consult with Mr. 
Roebling. Kaluf was favorably im- 
pressed, and on this basis the Equipment 
Board reported to the Commandant that 
". . . subject boat has possibilities for use 
in landing troops and supplies at points 
not accessible to other types of small 
boats." In May 1938 the Commandant 
cited this opinion in recommending to the 
Navy that "... steps be taken to procure 
a pilot model of this type of amphibious 
boat for further tests under service condi- 
tions and during Fleet Landing Exercise 
No. 5." 3« 

Both the Navy Board and the Bureau 
of Construction and Repair endorsed the 
recommendation unfavorably on the 
grounds of economy. The boat develop- 

^Itid., 163. 

^Linsert Interview; LtCol V. J. Croizat, "The 
Marines' Amphibian," MC Gazette, June 1953, 
42-43 (Croizat takes his information from 

^^ CMC Itr to Senior Member, NavDept ContBd 
for Dev of Landing Boats, 18May38, War Plans 
Files, 1924-39. 



ment program was at last well under way, 
and it seemed unwise to divert any of the 
limited appropriations to a purely experi- 
mental project. CNO concurred in the 
recommendation of the Board.^' 

Marine interest in the amphibian tractor 
persisted, however, and in October 1939, 
General Moses visited Roebling at his shop 
in Clearwater, Florida. He inspected the 
latest model tractor, and persuaded Roe- 
bling to design a model including desired 
military characteristics.^® 

In January 1940, Roebling had com- 
pleted the new design. An appropriation 
was secured from the Bureau of Ships, 
and work started on the first military 
model of an amiDhibian tractor. In No- 
vember the completed machine was 
delivered at Quantico where it was demon- 
strated for the Commandant and a large 
party of high ranking officers of the Army 
and Navy.^** It measured up in every 
respect save one. Its aluminum construc- 
tion was not considered rugged enough 
for hard military use. Still the tractor 
was so impressive in every other respect 
that the Navy contracted with Roebling 
for 200 of the machines constructed of 
steel. As Roebling did not have the facili- 
ties for mass manufacture, he subcon- 
tracted the actual construction to the 
Food Machinery Corporation which had 
a plant in nearby Dunedin. The first ve- 
hicle, now designated LVT(l) (Landing 
Vehicle Tracked), came off the assembly 
line in July 1941.*" 

^' Hid., and endorsements thereto. 

^ Pres MCEB Itr to CMC, 29Aug40, 2455-130- 
20 HQMC files; Linsert Interview; Croizat, op. 

"Linsert Interview ; Croizat, op. cit. 

" Chief BuShips Itr to Cdt 5th Naval Dist., 
6Dec40, 2455-130-60 HQMC files; SecNav 
ContBd for the Dev of Landing Vehicle, 

Quantity procurement of LVT(l) did 
not halt further development of amphi- 
bian tractors. By October 1941, the 
prototype of LVT(2) had put in an ap- 
pearance, but volume production of the 
new model was delayed by the entry of 
the United States into the war. To 
achieve maximum output, the design of 
LVT(l) was "frozen" shortly after Pearl 
Harbor and the vehicle put into mass 

This early LVT(l) was unarmed, 
though capable of mounting machine 
guns. The Marines, now that they had 
made a start, wanted something more : an 
armored, turreted model capable of 
mounting at least a 37mm gun and serving 
as the equivalent of a seagoing tank in 
landing operations. At Clearwater in 
January 1940, Roebling sketched a tur- 
reted version of the LVT) the plans for 
which Major Linsert, Secretary of the 
Equipment Board, later completed.*- 

Nothing more was done about the 
armored LVT until June 1941, when the 
Commandant recommended that such a 
vehicle be developed, using the existing 
LVT as a basis. The new vehicle should 
be " . . . capable of sustained point-blank 
combat against shore-based weapons .... 
It should be able to approach a defended 
beach from the sea, land, over-run enemy 
weapons, destroy them, and continue op- 
erations ashore to support our ground 
troops." *^ Armor protection against .50 
caliber machine-gun fire and an armament 

Tracked, "History of Landing Vehicle Tracked," 
lDec45, hereinafter cited as LVT Hist.; Daggett 

" LVT Hist. 

" Croizat, op. cit. 

" CMC Itr to CNO, 27Jun41, and CNO 1st en- 
dorsement thereto to Chief BuShips, 15Jul41, 
2455-130-20 HQMC files. 



including a 37mm antitank gun and three 
.30 caliber machine guns would be re- 
quired to accomplish this mission. The 
Chief of Naval Operations approved the 
project and directed the Bureau of Ships 
to perfect a design. 

Bureau engineers began development in 
cooperation with Eoebling and the engi- 
neers of the Food Machinery Corporation. 
But theirs was not to be the first armored 
LVT completed. Working independently 
and at its own expense, the Borg- Warner 
Corporation produced model "A," the first 
turreted amphibian tractor. Design work 
on the Roebling-Food Machinery model, 
LVT (A) (1) was not completed until De- 
cember 1941, and the prototype did not 
emerge from the Food Machinery plant 
until June 1942. It was an LVT (2) hull 
mounting a 37mm gun in a standard light 
tank turret. It Avas quickly put in pro- 

duction, and the first vehicle rolled off the 
assembly line in August 1943.** 

The craft described here were, of course, 
only a few of the wide variety of boats 
and beaching ships that performed yeo- 
man service in all theaters during World 
War II. These ranged in size from the 
big lumbering LST (Landing Ship, 
Tanks, or "Large, Slow Target"), orig- 
inated by the British, to the Army-devel- 
oped DUKW, an amphibious truck 
propeller-driven afloat. But Marines 
played no notable part in the development 
of any of these, and none had appeared 
during the period covered by this volume. 
They will be described in subsequent vol- 
umes as they came to play their part in 
the tactical picture of Marine operations. 

'LVT Hist; Croizat, op. cit. 


Marine Occupation of Iceland' 

"It has been said," wrote Winston 
Churchill, " 'Whoever possesses Iceland 
holds a pistol firmly pointed at England, 
America, and Canada.' " ^ At the time of 
which he wrote, the "pointed pistol" 
threatened most immediately the British 
lifeline : the northern convoy route be- 
tween Great Britain and the Western 
Hemisphere, upon which the island king- 
dom was dependent for most of the mate- 
rials to sustain its war effort as well as 
much that was needed for its very sub- 

^ Unless otherwise noted the material in this 
chapter is derived from the 1st MarBrig(Prov) 
Rept of Activities 16Jun41-25Mar42, 26Mar42 ; 
6th Mar(Reinf) Repts of Activities 25May- 
30Nov41, 13Dec41 ; 5th DefBn Repts of Activities 
7Jun41-28Feb42, 27Feb42 ; Correspondence files 
dealing with Marine occupation of Iceland ; 
J. L. Zimmerman, Notes and MSS on Marine 
occupation of Iceland (located at NRMC, Job 
14051, Box 9, Folders 129-130), hereinafter cited 
as Zimmerman MSS; Gen O. P. Smith, Diary 
and Narrative covering the occupation of Ice- 
land, hereinafter cited as Smith Narrative; 
S. Conn and B. Fairchild, "The Framework of 
Hemisphere Defense," MS of a forthcoming 
volume in the series United States Army in World 
War II (located at OCMH), hereinafter cited as 
Hemisphere Defense; B. Fairchild, MS chapters 
titled "Planning the Iceland Operation: The 
Army's First Task Force," "Establishing the Ice- 
land Base Command," and "Bermuda and the 
North Atlantic Bases," part of a forthcoming 
volume of the same series ; W. L. Langer and S. E. 
Gleason, The Undeclared War (New York: 
Harper & Brothers, 1953), hereinafter cited as 
Undeclared War. 

'W. S. Churchill, The Grand Alliance (Bos- 
ton: Houghton Mifflin Company, 1950), 138. 

sistence. Iceland perched on the flank of 
these shipping lanes, which were under 
heavy attack by German submarines. 
Hostile air and naval bases on the island 
would almost certainly render the north- 
ern route unusable, and put pressure, per- 
haps intolerable pressure, on the longer 
and more vulnerable southern route. 

At the outbreak of the war Iceland en- 
joyed the status of autonomous parlia- 
mentary monarchy, sharing the same king 
with Denmark. When the Nazis overran 
the latter nation in April 1940, the Ice- 
landic Parliament voted to take over the 
executive power of the Danish King and 
to assume control of foreign affairs. The 
strategic island became, for all practical 
purposes, a completely independent re- 
public ^ — and a wholly defenseless one 
without even the pretense of an army or 
navy. This state of affairs gave rise to 
considerable concern in London and 
Washington, more genuine concern than 
it caused initially among the insular- 
minded Icelanders. 

To the British the threat appeared very 
desperate indeed. Early in May they de- 
termined to occupy Iceland, and the need 
for speed and secrecy fused decision and 
action.* There was no time to stand on 

^ On 16May42 the Parliament announced that 
Iceland would not renew its union with Den- 
mark and in 1944 the island became in name as 
well as fact a republic. 

* J. R. M. Butler, Grand Strategy: Volume 
II — History of the Second World War (London: 
HMSO, 1957), 262. 




ceremony; despite Churchill's bland as- 
sertion that the British occupation of Ice- 
land was effected "with the concurrence 
of its people," ^ they had, in fact, not 
been consulted beforehand. "As the atti- 
tude likely to be adopted by the Icelandic 
Government toward such an 'invasion' 
was in some doubt they were not informed 
of the proposed expedition." " Indeed the 
first inkling the natives had that anything 
out of the ordinary was afoot came when 
early-rising fishermen discovered a Brit- 
ish destroyer nosing up to a jetty in the 
harbor of the island capital, Reykjavik. 
At 0620 on 10 May, a reinforced battalion 
of Royal Marines landed and occupied the 
town, moving so swiftly that it was able to 
seize the German Consulate before the 
hapless Consul could destroy his papers. 

According to plan, the Royal Marines 
were to take the situation in hand in or- 
der to pave the way for larger occupa- 
tion forces. They were relieved in ten 
days by a Canadian Ai'my brigade which 
was first i-einforced and later replaced 
by British units. By the time Iceland 
began to loom large in U. S. defense 
plans, the big, bleak, sparsely-populated 
island was occupied by nearly 25,000 Bri- 
tish troops. Hvalf jordur, a deep inlet of 
the sea 30 miles north of Reykjavik, be- 
came the site of a vital naval fueling and 
repair base, while the principal airfields, 
also near the capital, were home bases for 
squadrons of patrol bombers that hunted 
the German submarines.^ 

As reverse followed reverse, however, 
the British increasingly felt the need for 
the return of their troops from Iceland 

to the home islands, seriously threatened 
with invasion and under heavy air at- 
tack. The prospect of British withdrawal 
caused some alarm among the Icelanders 
and led to diplomatic soundings of the 
American position. 

On 18 December 1940 the Icelandic 
Minister of Foreign Affairs, V. Stefans- 
son, arranged a private meeting with the 
U. S. Consul General, Bertel E. Kuni- 
holm. After firm assurances that his 
proposal was strictly unofficial, the Min- 
ister suggested to Kuniholm that the 
United States might consider the possi- 
bility of declaring Iceland part of the 
area covered by the Monroe Doctrine, in 
effect joining the island to the Western 
Hemisphere.® Kuniholm duly reported 
the tentative proposition to Washington 
and nearly a month later he received a 
cautious reply from the Secretary of State 
which advised him that no action was 
likely to be forthcoming in the near fu- 
ture but that he should neither encour- 
age nor discourage further approaches 
along this line.® 

In unheralded American-British staff 
conversations which took place in Wash- 
ington in the first months of 1941, plans 
were laid for Allied action in case the 
U. S. should be drawn into the war be- 
side Britain. Under these plans the de- 
fense of Iceland was to become the re- 

° Churchill, loc. cit. 

"Maj D. B. Drysdale, RM, Itr to LtCol J. L. 
Zimmerman, 7Sep54, in Zimmerman MS8, 
Folder 130. 

' Butler, op. cit., 262, 287, 402, 469. 

* Although the location of the eastern bound- 
ary of the Western Hemisphere is a subject of 
debate among geographers, most mai>s of this 
period show Iceland as clearly within the East- 
ern Hemisphere. Secretary Hull, however, re- 
membered associates bringing him maps (at the 
time Hitler seized Denmark) which showed 
Greenland wholly and Iceland partly within the 
Western Hemisphere. The Memoirs of Cordell 
Hull, 2 vols ( New York : The Macmillan Com- 
pany, 1948), I, 73. 

" Ibid., I, 754. 



sponsibility of the United States; Army 
troops were to relieve the British as soon 
as practicable after the outbreak of war, 
but certainly no sooner than 1 September 
1941, as the Army did not feel it would 
be ready to take on such a commitment 
until then.^" But as the spring of 1941 
wore on, American measures in aid of 
Britain, such as Lend-Ijease and the pro- 
gressive extension of the Neutrality Pa- 
trol into the mid-Atlantic, brought the 
U. S. closer and closer to conflict with 
Germany. Open and increasing support 
of the British seemed to suit the public 
mood; a survey of public opinion taken 
by the Gallup Poll in early May showed 
that an overwhelming majority (75%) of 
the American people favored helping Brit- 
ain even if such a course was sure to lead 
the nation into war with Germany." The 
stage was thus set for what one exhaustive 
study of this period has called an "overt 
act of participation in the European 
conflict." " 

By late spring Britain felt her back 
against the wall. Churchill asked Presi- 
dent Koosevelt to send American troops 
to Iceland to replace the British garrison. 
The President agreed provided an invita- 
tion to the American occupation force was 
forthcoming from the Icelandic Govern- 
ment. Churchill undertook to produce 
this invitation, but the process proved 
more one of extraction than of produc- 
tion. Icelandic reluctance to "invite" a 

"M. E. Matloff and E. M. Snell, Strategic 
Planning for Coalition Warfare — United States 
Army in World War II (Washington: OCMH, 
1953), 46, hereinafter cited as Strategic Plan- 

"E. Roosevelt and J. P. Lash (eds.), F. D. R.: 
His Personal Letters 1928-191,5, 2 vols (New 
York: Duell, Sloan and Pearce, 1950), II, 1158. 

" Langer and Gleason, op. cit., 575. 

foreign force to occupy the island very 
nearly upset a timetable already in oper- 

On 4 June, the President ordered the 
Army to prepare a plan for the immedi- 
ate relief of British troops in Iceland. 
The question of where the troops were go- 
ing to come from arose immediately. Al- 
though the Army had reached a strength 
of nearly a million and a half men, the 
great bulk of its soldiers were raw re- 
cruits gathered in by Selective Service 
and recently called up National Guards- 
men. Under existing legislation these 
men could not be sent beyond the West- 
ern Hemisphere unless they volunteered 
for such service. Equipment in nearly 
every category was in short supply, even 
for training purposes. The Army needed 
its comparatively small force of regulars 
to form cadres for new units. To with- 
draw these cadres for an expeditionary 
force would throw the whole immense 
training program out of gear. 

A review of the Army's immediate capa- 
bilities convinced the President that the 
Marine Corps would have to furnish the 
initial occupation force for Iceland. 
Since all Marines, both regular and re- 
serve, were volunteers, there were no geo- 
graphical restrictions on their use. On 
5 June, Roosevelt directed the Chief of 
Naval Operations (CNO), Admiral 
Harold R. Stark, to have a Marine bri- 
gade ready to sail in 15 days' time. The 
organization of this brigade was facili- 
tated by the fact that a reinforced infan- 
try regiment slated for expeditionary duty 
was at that moment en route from the 
west coast to the east. 

At this time the Marine Corps was 
heavily committed to a program of organ- 
izing, equipping, and training two divi- 
sions, one on each coast. Since the infan- 



try regiments of both divisions were still 
forming, they were considerably under- 
strength, and it had been necessary to 
reinforce the east coast's 1st Marine 
Division when it was tabbed for a major 
role in a proposed landing operation. On 
24 May, the Commandant drew on the 2d 
Marine Division at Camp Elliott, Cali- 
fornia, for the necessary regiment, and 
Colonel Leo D. Hermle's 6th Marines 
(Reinforced) was selected "for temporary 
shore duty beyond the seas." ^^ The regi- 
ment was brought up to full strength 
by substantial drafts from the 2d and 8th 
Marines," and on 28 May it joined its 
assigned reinforcing artillery, tank, and 
service elements. Six days aft«r he re- 
ceived his orders, Colonel Hermle had his 
command combat loaded; the ships, three 
large transports and four destroyer trans- 
ports, sailed from San Diego on 31 May. 

When it had embarked, this regiment 
had orders to report to the Commanding 
General, I Corps (Provisional), FMF, 
Atlantic Fleet. At that time, its most 
probable mission appeared to be either the 
seizure of Martinique or the occupation of 
the Azores, both discussed in the follow- 
ing chapter. Momentous events, however, 
were developing in Europe, and these 
served to change the whole pattern of the 
war, as well as the mission of the regiment. 
Both British and American intelligence 
indicated that Hitler was getting ready 
to attack Russia, and soon. Such an 
event would automatically cancel any im- 
mediate threat to Gibraltar and render 
the Azores venture pointless. President 

"6th Mar(Reinf)Repts, op. cit., 1. 

" "The rule was that [these] men must have 
been in the service for one year and must have 
clear records. The other regiments 'played ball' 
in this respect and we received good men." 
Smith Narrative, 17. 

Roosevelt, in fact, ordered a suspension of 
planning for the Azores operation on 7 
June, while preparations for the move- 
ment to Iceland proceeded apace. 

While the 6th Marines' convoy was still 
in the Pacific heading for the Panama 
Canal, the wheels were set in motion to 
complete the organization of the pro- 
jected brigade. One other major unit, the 
5th Defense Battalion at Parris Island, 
was designated for duty in Iceland; its 
commanding officer, Colonel Lloyd L. 
Leech, flew to Washington on 7 June for 
a two-day round of briefing and reports. 
The battalion's antiaircraft guns and 
gunners were what was wanted, so when 
the order assigning the 5th Defense to 
I Corps (Provisional) was published on 
10 June the 5-inch Artillery Group was 
shown as being detached. In addition to 
the 6th Marines (Reinforced) and the 5th 
Defense Battalion (less 5-inch Artillery 
Group), the budding brigade received a 
company of engineers, a chemical platoon, 
and a platoon of scout cars from the 1st 
Marine Division at New River. The port 
for the hurried assembly of ships, mate- 
riel, and men was Charleston, S. C. 

The men of the 5th Defense Battalion 
had some inkling of their probable area 
of employment; Colonel Leech's warning 
order phoned from Washington on the 
8th had directed that special attention 
be paid to provision of warm clothing. 
On board the 6th Marines' transports, 
however, speculation was rife that the 
regiment was heading for the Caribbean, 
perhaps for Guantanamo Bay, but more 
popular was the rumored destination of 
Martinique.. When the convoy turned 
north after clearing the canal, passed the 
western end of Cuba, and headed for 
Charleston most of the "scuttlebutt" still 
held out for a tropical objective. Need- 



less to say, the issue of winter clothing 
after the regiment arrived at Charleston 
on 15 June came as a real "shocker." The 
severely limited time to assemble and load 
out the Iceland force made this cold 
weather gear "the darndest collection of 
winter clothing ever assembled;" ^^ there 
were bits and pieces of everything. 

On the day following the arrival of the 
6th Marines in Charleston the 1st Marine 
Brigade (Provisional) was formally or- 
ganized; its commander was Brigadier 
General John Marston. The troop list 
included : 

Brigade Headquarters Platoon 

Brigade Band 

6th Marines 

5th Defense Battalion (less 5-inch Artillery 

2d Battalion, 10th Marines 

Company A, 2d Tank Battalion (less 3d Pla- 

Company A, 2d Medical Battalion 

Company C, 1st Engineer Battalion 

1st Platoon, Company A, 2d Service Battalion 

3d Platoon, 1st Scout Company 

Chemical Platoon 

On 18 June, General Marston arrived in 
Charleston from Quantico, bringing with 
him a small headquarters detachment and 
his instructions from the CNO for the op- 
eration of his brigade in Iceland. These 
orders, dated 16 June, gave liim a simple 
and direct mission : 

In Cooperation with the British Garrison, De- 
fend Iceland Against Hostile Attack.'" 

The question of over-all command in Ice- 
land had, of course, risen early in the top- 
level negotiations. The British wished 
the brigade to be placed directly under 
their control since they had tlie major 

force on the island, but Admiral Stark 
thought that it would be going too far for 
U. S. troops, ostensibly neutral, to be 
placed under the command of an officer of 
a belligerent power. Marston's orders, 
therefore, read that he would coordinate 
his actions "with the defense operations of 
the British by the method of mutual co- 
operation," " while reporting directly to 
the CNO. 

The brigade spent a week in Charleston, 
most of it devoted to loading supplies 
that arrived from camps and depots all 
over the eastern half of the U. S. The 
Army might not be sending any troops 
in this first contingent, but a good portion 
of the weapons and equipment that went 
out with the Marines was taken from 
Army units." On 22 June, the last cargo 
that could be handled within the time 
limits set was loaded and at 0800 the four 
transports and two cargo vessels carrying 
4,095 officers and men set sail for Argentia, 

At sea a formidable escort force includ- 
ing battleships, a couple of cruisers, and 
ten destroyers joined up.^® Five days out 
of Charleston, the convoy arrived at 
Argentia and hove to awaiting further 
orders. These orders were not forthcom- 
ing until 1 July, when the Icelandic reluc- 
tance to actually "invite" American occu- 
pation was finally compromised in a much- 
qualified statement by the island's Prime 
Minister to President Roosevelt that the 
presence of U. S. troops was "in accord- 
ance with the interest of Iceland." ^° This 
left-handed invitation was the go-ahead 

''MajGen H. R. Paige Itr to ACofS, G-3, 
HQMC, February 1957. 

16 CNO Serial 069312 to CG, 1st MarBrig- 
(Prov), 16Jun41. 

18G-4 draft memo for TAG, "Transfers of 
Equipment to the U. S. Marine Corps," .5Jun41. 

'' U. S. Atlantic Fit OPlan F-41, 20Jun41, 1-2. 

* Msg sent by Prime Minister Herman Jonas- 
son of Iceland to President Roosevelt, lJul41. 



signal and the brigade was headed east 
by dawn on 2 July. The Marines were go- 
ing with the blessing of Churchill who 
had written the President earlier that: 

I am much encouraged by . . . your marines 
taking over that cold place and I hope that once 
the first installment has arrived you will give 
full publicity to it. It would give us hope to 
face the long haul that lies ahead.^ 

The President made the desired an- 
nouncement on 7 July as the convoy an- 
chored in Reykjavik harbor, pointing out 
that the Americans were there "to supple- 
ment, and eventually to replace, the Brit- 
ish forces," and that an adequate defense 
of the strategic island was necessary to 
ward off a potential threat to the Western 
Hemisphere.^^ A third, but unannounced, 
purpose of this American occupation was 
the acquisition of a naval and air base in 
Iceland to facilitate the prosecution of our 
antisubmarine war in the North Atlantic.^^ 

While the threat of German attack was 
always present, the likelihood of it hap- 
pening steadily lessened as the year wore 
on.^* On the day that the 1st Brigade left 

" Quoted in Hull Memoirs, op. cit., II, 947. 

" S. I. Rosenman (ed). The Public Papers and 
Addresses of Franklin D. Roosevelt, 13 vols 
(New York: Harper and Brothers, 1950), X, 

'" S. E. Morison, The Battle of the Atlantic — 
September 1939-May 1943— History of the 
United States Naval Operations in World War 
II (Boston: Little, Brown and Company, 1947), 
78, hereinafter cited as Battle of the Atlantic. 

" An estimate of the situation prepared by a 
special beard convened by the brigade shortly 
after its arrival in Iceland attributed to the 
Germans the following capabilities : To land in 
force from air or sea ; to conduct bombing at- 
tacks ; and to conduct raids by surface vessel 
and submarines. The board concluded, however, 
that as long as the British Home Fleet oper- 
ated in superior numbers in the water surround- 
ing Northern Scotland, the Orkney, Shetland, 

Charleston, Grermany attacked Russia. 
Hitler repeatedly in the months that fol- 
lowed indicated that he wanted to avoid 
provoking the U. S. into war while he 
concentrated on the offensive in Russia. 
His submarine commanders were given 
orders to spare American shipping as 
much as possible, even though it had been 
publicly announced that U. S. Navy ves- 
sels were affording protection to British 
and Canadian ships that joined Ameri- 
can convoys headed for Iceland. Still 
Hitler decreed that there would be no ac- 
counting for the submarine commander 
who sank an American vessel by mistake. 
Up until the actual U. S. entry into the 
war this partial immunity of American 
vessels from attack held good.^^ 

The fact that Hitler had decided to go 
easy on U. S. ships in the North Atlantic 
was naturally not known to American 
naval commanders. There was consider- 
able pressure to get the brigade and its 
equipment unloaded in the shortest pos- 
sible time and the convoy headed back for 
the States. This unloading proved an 
onerous task. There was little local labor. 
Marines had to furnish all working parties 
and the men toiled around the clock, 
helped not a little by the fact that at this 
time of year it was light 24 hours a day. 

Only two ships could be docked at Rey- 
kjavik at a time and the places beside 

and Faroe Islands it would be impossible for 
the Germans to support a force of any size in 
Iceland. IstMarBrig(Prov) Estimate of the Sit- 
uation (Defense of Iceland), 5Aug41. 

-' "Fuehrer Conferences on Naval Affairs, 
1939-1945," Brassey's Naval Annual 1948 (New 
York: The Macmillan Company, 1948), 220j5f. 
See the transcripts for the conferences of 
21Jun41, 9Jul41, 25Jul41, 17Sep41, and 13Nov41 
for the continuity of German policy regarding 
American shipping. 



the wharves were reserved for the cargo 
vessels which carried heavy equipment of 
the 5th Defense Battalion. The rest of 
the convoy rode at anchor in the harbor, 
while men and supplies were lightered 
ashore to a gently sloping pebble beach 
near the city. Early on 12 July the job 
was finished, the convoy sailed, and the 
Marines had their first real chance to look 
around them. 

They drew small reassurance from what 
they saw. The Icelandic landscape was 
something less than prepossessing, at least 
to men raised where soil produces vegeta- 
tion and a tree is a tree. No trees above 
dwarf height grow on Iceland's rugged, 
mountainous terrain, and vegetation is 
limited to a little sheep pasturage on the 
comparatively flat stretches. It has been 
described as the most volcanic region in 
the world. Craters, many of them oc- 
casionally active, pock its surface, and 
lava flows lace across it. 

The most unpleasant thing about Ice- 
land's weather is its very uncertainty ; the 
mountains usually insure that the same 
kind of weather rarely exists simulta- 
neously all over the island. Although the 
temperature range is moderate, the hu- 
midity is consistently high, and precipita- 
tion frequent but erratic. About the only 
constant is the assurance of steady winds, 
which may change abruptly to gale 

The island is slightly smaller in area 
than Kentucky, but barely supported a 
population of about 120,000 at the time 
of the occupation. Along its 2,300 miles 

of jagged coastline were a number of 
small fishing villages; and except for the 
area around Reykjavik where there was 
a roadnet, all communication was by sea. 
The prim little capital boasted about 
38,000 inhabitants, two movie houses, and 
one first class hotel; as a liberty town for 
nearly 30,000 British and American troops 
it boasted nothing. The only living 
things the island had in abundance were 
sheep and ponies," and the Marines never 
developed a taste for mutton and were 
forbidden to ride the runt-sized steeds. 
Altogether, it was probably good for 
morale that the Marines did not know at 
this time that they were destined to see 
Iceland — and nothing but Iceland — for 
eight dreary months to come. 

Even before the first brigade unit set 
foot on shore, the Marines learned what 
the term "mutual cooperation" meant to 
the British. They could not have been 
more cordial, generous, and helpful. As 
the brigade was woefully short of motor 
transport, the British put more than 50 
trucks at its disposal, together with driv- 
ers familiar with the region and the traffic 
problems peculiar to Iceland — and left 
them in the hands of the Marines for sev- 
eral weeks. They also furnished rations 
and turned over several of their perma- 
nent camps to the new arrivals, moving 
into tent camps to make room.^® 

The enthusiastic reception by the Brit- 
ish included a highly prized offer by their 

"* In a hurricane on 15Jan42, wind velocities 
of over 125 mph were recorded. It did an enor- 
mous amount of damage. Ships were driven on 
the rocks and huts and other buildings which 
were not firmly anchored were blown away. 
Paige, op. cit. 

" Most of the information on Iceland's climate 
and terrain was taken from Col L. P. Hunt, "Re- 
port of two-day reconnaissance of Iceland, June 
12-13, 1941," 18Jun41. 

" "Our reception by the British has been 
splendid. They have placed at our disposal all 
of their equipment and have rationed us for ten 
(10) days to cover the period of disembarka- 
tion." BriGen J. Marston Itr to MGC, llJul41. 



commander, Major General H. O. Curtis, 
to provide the Marines with the distinc- 
tive polar bear shoulder insignia of the 
British force. General Marston accepted 
for the brigade and noted later that : 

The mutual cooperation directive worked, to 
the entire satisfaction of the British Com- 
mander and the Brigade. The British complied 
with our requests and we complied with theirs. 
It was as simple as that. A British commander 
less sympathetic than General Curtis might have 
upset the applecart but under that talented offi- 
cer no incident of conflict occurred.'* 

In their new camps the Marines made 
their first acquaintance with the Nissen 
hut, an introduction that was to ripen 
into familiarity that rarely reached the 
friendship stage. In the months to come 
the men of the brigade were to build and 
maintain roads and construct defenses; 
they were to become very practiced at the 
art of the stevedore; but most of all they 
were to become efficient builders of the 
ubiquitous Nissen hut. The hut itself 
"was an elongated igloo covered with cor- 
rugated iron roofing and lined with 
beaver board" ^^ designed to accommo- 
date about 14 men. It was possible to 
erect several huts in combination to ac- 
commodate larger numbers of men or for 
use as offices, mess halls, recreation rooms, 
and classrooms. 

For the first week ashore the Marines 
were fully occupied getting their camps 
established and then they were fitted into 
the British scheme of defense. Initially, 
the brigade's primary mission was to 
serve as a mobile reserve although its lack 
of transportation meant that most of its 
mobility would be dependent on foot 

power.^^ The various units, which were 
spread out over a good part of the coun- 
tryside around Reykjavik, were also re- 
sponsible for local defense of their biv- 
ouac areas, a responsibility that grew to 
include long segments of coastline when 
the British units defending these possible 
landing points were later relieved. 

The machine guns and 3-inch guns of 
the 5th Defense Battalion were integrated 
into the British antiaircraft defenses 
around the airfield and harbor and re- 
mained a part of this system for the rest 
of the Marines' stay. As a result, the 5th 
Defense spent most of its time performing 
the duties for which it was constituted; 
its state of training was good and it im- 
proved as a result of a steady round of 
gun watches and drills and frequent 
though unproductive enemy aircraft 
alerts. In contrast, the men of the 6th 
Marines and its reinforcing units had 
reason to think that they were on one 
gigantic and never-ending working party, 
and the regiment labelled itself a "labor 
regiment" in its August report to General 

A welcome break from the steady grind 
of labor details occurred on 16 August 
when Prime Minister Churchill visited 
Iceland en route to England following his 
famous Atlantic conference with Presi- 
dent Roosevelt. He was accompanied by 
an imposing array of high British rank: 
Admiral of the Fleet Sir Dudley Pound, 
First Sea Lord ; General John Dill, Chief 
of the Imperial General Staff; and Air 
Chief Marshal Sir Wilfred Freeman, 
Vice Chief of Air Staff. After paying 
their respects to local officials, they at- 

" MajGen J. Marston Itr to ACof S, G-3, HQMC, 

" Smith Narrative, 34. 

'' Iceland Force memo IF/168/1/G to CG, 1st 
MarBrig, 16Jul41 ; 1st MarBrig OpOrd No 3^1, 



tended a large joint British-American 
military review held in their honor. Of 
this event Churchill wrote later: "There 
was a long march past in threes, during 
which the time 'United States Marines' 
bit so deeply into my memory that I could 
not get it out of my head." ^^ 

The reason for the continuous round 
of camp construction was two-fold. 
First, somebody had to build the camps 
to accommodate the expected influx of 
Army troops; neither the British nor the 
Icelanders were in a position to do so. 
The process of simple elimination gave 
the Marines the job. Second, it soon be- 
came apparent that the Marines them- 
selves were going to stay for a while and 
a good part of their time had to be spent 
preparing their own facilities for the on- 
set of winter. 

A common, indeed, official, belief that 
the Marines were going to be relieved in 
September by Army troops held strongly 
for about a month after the brigade ar- 
rived in Iceland. There were numerous 
evidences that this was the intention of 
the top planners when the concept of the 
Marine Corps furnishing the initial oc- 
cupation troops was first broached. By 
mid-August, however, it became evident 
that the Army would not be able to pro- 
vide enough men to relieve the brigade 
and that the lack of readily available 
troops would make the role of those who 
did arrive one of reinforcement rather 
than relief. The British, who were sup- 
posed to return to their home islands, had 
to stay on to bolster the defenses. The 
crux of the Army's dilemma was the fact 
that not all of its men were available for 
assignment; "the passage of legislation 
in August 1941 permitting the retention 

in service of the selectees. Reserve officers, 
and the National Guardsmen still left the 
problem of restriction on territorial serv- 
ice — a problem which was to remain with 
the Army until Pearl Harbor brought a 
declaration of war." ^^ 

There was really not too much trouble 
taking care of the first Army contingent 
to ari-ive, a small force of about 1,000 
men built around a pursuit squadron and 
an engineer battalion. Their convoy 
made port on 6 August and the units, 
which came under Marston's command, 
moved into a camp set up for them by 
the Marines. However, preparations for 
the arrival of a second Army echelon of 
brigade strength due in mid-September 
meant that every Marine available had to 
turn to on camp construction. It was 
the difficulties attendant upon the raising 
of this second force that led to the de- 
cision to hold the Marines in Iceland.^* 

The commander of the Army troops of 
the September echelon was senior to Gen- 
eral Marston; according to the original 
occupation plan, the principle of unity of 
command was to hold in Iceland, and 
under it the senior officer present, regard- 
less of service of origin, would have as- 
sumed operational control over all Ameri- 
can troops. Acording to this concept, 
Army Major General Charles H. Bone- 
steel would simply have superseded Gen- 
eral Marston and all hands would have 
carried on as before. But in the interim 
between June and September, the Army 
Chief of Staff, General George C. Mar- 

' Churchill, op. cit., 449. 

" Statcgic Planning, 51. 

^ AG memo to ACof S, War Plans Dlv, 6Sep41. 
In order to field the force that finally reached 
Iceland in September, the Army had to draw 
on posts and stations all over the U. S. AG 
WrnO to Army commanders concerned, 14Aug41 
(located at TAGO). 



shall, had decided that unity of command 
did not go far enough, at least as far as 
Iceland was concerned. He determined 
that if General Bonesteel was to have 
full responsibility for the American occu- 
pation, then he should also have full ad- 
ministrative as well as operational control 
over all the troops in Iceland. 

Such a transfer of the Marines from 
Navy control could be effected by execu- 
tive order, as had been done by President 
Wilson in the case of the Marines serving 
in France in World War I. Unfortu- 
nately, from the Marines' point of view, 
this transfer involved a great deal more 
than a simple change of command. It 
brought them under the Army's adminis- 
trative and disciplinary system which dif- 
fered considerably from that of the Navy 
and with which they were unfamiliar. 

The Commandant, who had seen the 
system at work in World War I, protested 
vigorously. On 4 September he wrote 
Admiral Stark : 

The proposed change will not only necessitate 
a complete revision of this plan [unity of com- 
mand] but would introduce many administrative 
diflBeulties, with no corresponding advantages 
in so far as command relations are concerned. 
A complete change of the administrative system 
would again be required when the First Marine 
Brigade is detached from the Army.** 

And again on 5 September : 

In view of the existing situation in Iceland 
and the probable nature of other operations to 
be conducted by the Navy elsewhere, the pro- 
posed plan has many undesirable ramifications. 
If carried to its logical conclusion, it will mean, 
at best, frequently shifting Marine units from 
the Navy to the Army and back again, with 
much administrative grief. It will probably 
change our concept of command relations in 
joint operations.™ 

But it was a losing fight. Marshall 
stated that he had no intention of estab- 
lishing a precedent and remained ada- 
mant. The Commandant did not learn of 
the proposed change until it was prac- 
tically an accomplished fact, and the sup- 
port he received from the CNO was luke- 
warm. The actual transfer of command 
took place on 24 September and General 
Holcomb was directed to report to the 
Secretary of War on all matters pertain- 
ing to the brigade.^^ 

The resultant administrative difficulties 
did not prove to be as bad as Holcomb and 
many others had feared. The change- 
over was more of an annoyance than it 
was a definite hindrance; after all, as one 
battalion commander commented later, 
"while administration difficulties may be 
bothersome they can be handled."^® In 
the course of trying to master Army pro- 
cedures, General Marston wrote the As- 
istant Commandant: 

They have a tremendous amount of paper work 
which the Marine Corps seems able to avoid. 
The barrage of force orders coming out of staff 
sections is appalling. Of course we are getting 
along all right but it will be months before we 
are oriented in the new direction ... If the 
future develops another situation similar to that 
of this Brigade in Iceland, I hope that you will 
be able to have the transfer deferred with at 
least two months notice so that the oflScers con- 
cerned can get themselves oriented in prepara- 
tion for the jurnp.'^ 

One of General Bonesteel's first acts as 
the Commanding General of the new Ice- 
land Base Command was to send a letter 
of appreciation to the 1st Marine Brigade 
(Provisional) which extended his "sin- 

" MGC memo for Adm Stark, 4Sept41. 
'' MGC memo for Adm Stark, 5Sep41. 

" Presidential directive to SecWar and SecNav, 

''MajGen W. A. Worton Itr to CMC, lFeb57. 

™BriGen J. Marston Itr to BriGen A. A. 
Vandegrift, 100ct41. 



cere thanks for the splendid assistance 
[given] in the preparation of the various 
campsites and in numerous other ways 
prior to and during our arrival in Ice- 
land. The amount of hard and extended 
labor involved is fully recognized and 
deeply appreciated." *° 

The onrush of winter made it necessary 
for all troops to devote a good part of 
their time to camp maintenance and 
weatherizing. And as supplies continued 
to come in for the depots being built up 
near Reykjavik, working parties had to 
be provided to empty ships as well as to 
construct the storehouses needed to pro- 
tect the equipment. Days rapidly short- 
ened until there were only four hours of 
a sort of hazy daylight to accomplish 
necessary functions. 

With the continued requirements for 
camp construction and preparations for 
an arctic winter, the brigade was not able 
to conduct a satisfactory training pro- 

Every possible opportunity was seized 
by unit commanders, however, to improve 
the state of readiness of their men. Many 
of the specialists, of course, like the 
communicators, engineers, and service 
personnel received considerable on-the-job 
training. While large-scale exercises 
were not possible, small units operated to- 
gether as the press of construction al- 
lowed. In particular, a considerable 
amount of range firing of crew-served 
weapons was accomplished. When the 3d 
Battalion of the 6th Marines was moved 
to a camp too far away from Reykjavik to 
make it feasible to use its men for working 
parties, the commanders of 1/6 and 2/6 
agreed to alternate in furnishing working 

parties "in order to get in a minimum 
amount of training." " The 3d Battalion, 
encamped in a pass that lay right in the 
path of winter winds howling out of the 
mountains near Hvalf jordur, was forced 
to "button-up" for the winter almost as 
soon as it shifted in September. 

The lack of adequate unit training has 
been emphasized by some critics of the 
Marines' employment in Iceland. Train- 
ing did not stop; it was hampered and 
curtailed by the weather and tlie require- 
ments of working details, but it did go on 
despite all the very real obstacles. The 
men, trained and indoctrinated as am- 
phibious assault troops, however, were 
perturbed when they heard the news of 
Pearl Harbor while huddled around the 
stoves in their Nissen huts. Were they 
to be left forgotten in the wrong ocean? 

Once the war broke out in earnest the 
Navy, too, did not view with favor the em- 
ployment of a Marine Brigade on a de- 
fensive mission in Iceland. The Marines 
were needed in the Pacific and pressure 
was put on the Army to get them relieved. 
Plans were laid to send a convoy with 
8,000 men from New York on 15 January 
to provide the brigade's relief and return 
transportation. But, like so many pre- 
vious false starts, this was not to be. Sev- 
eral of the ships in this convoy were 
diverted elsewhere and the resulting troop 
lift was only enough to relieve one battal- 
ion. General Marston picked 3/6, which 
cheerfully turned over its wind-blown 
billets to the Army troops and embarked 
on 28 January. The battalion left Ice- 
land on the 31st and reached Mew York 
on 11 February. 

A start had been made and the brigade 
began negotiations to turn over its camps, 

"CG, IBC Itr to CG, 1st MarBrig(Prov), 
27Sep41, quoted in Zdmmerman MSS, Folder 129. 

" Gen O. P. Smith Itr to ACof S, G-3, 7Feb57. 



defense mission, and heavy equipment to 
the Army. The convoy carrying the final 
relief put into Reykjavik on 3 March, and 
the Marines began loading out the follow- 
ing day. At 1010 on 8 March, General 
Mai'ston closed his CP on shore and 
opened it on board the USS McCawley\ 
at noon that date the brigade returned 
to the jurisdiction of the Navy. It is in- 
teresting to note that this is the only in- 
stance in World War II where a Marine 
unit was "detached for service with the 
Army by order of the President." In the 
many joint operations that followed, all 
services adhered to the principle of unity 
of command. General Bonesteel recog- 
nized the Marines' dislike for the "de- 
tached service" concept but in a final let- 
ter to General Marston commended the 
brigade whose "every officer and enlisted 
man gave his whole hearted support and 
cooperation to our efforts to a much 
greater extent than mere compliance with 
instructions implied." ^^ 

The brigade landed at New York on 
25 March and was immediately disbanded. 
The 5th Defense Battalion was ordered 
to Parris Island, the 6th Marines to the 
Second Division at Camp Elliott, and the 
supporting units to their parent organiza- 
tions wherever those might be. 

Thus passed into history an uncomfort- 
able and at times frustrating mission, the 
military value of which was not clearly 
apparent at the time. The Marine Corps' 
expansion program in late 1941 and early 
1942 was admittedly hampered by the ab- 
sence of such a sizeable body of well- 

"CG, IBC Itr to CG, 1st MarBrig(Prov), 
lMar42, quoted in Zimmerman M88, Folder 130. 

trained regulars and reserves. The bri- 
gade had relieved no appreciable number 
of British troops, which had been the orig- 
inal purpose of the American occupation. 
There is no concrete evidence that the 
Germans ever seriously considered attack- 
ing Iceland, although it is conceivable, 
even if somewhat unlikely, that the knowl- 
edge of the presence of the brigade might 
have deterred such an attack. The mili- 
tary value of the Iceland occupation 
stemmed from rigorous service in the field. 
In the many scattered and detached posts, 
heavy responsibilities fell on the shoulders 
of the young company grade officers 
and NCOs. Adversity developed and 
strengthened leadership. Once the bri- 
gade reached Iceland there was a mini- 
mum rotation of officers and men. This 
stability of personnel gave the command- 
ers an opportunity, seldom afforded in 
peacetime, to develop teamwork and unit 
esprit de corps. Upon return to the 
United States, almost all ranks received a 
promotion and all units of the brigade 
were drawn on heavily to provide leaders 
for newly activated units. The 6th Ma- 
rines furnished large drafts to the raider 
and parachute battalions, as well as to 
units of the 2d Division. 

The military know-how, discipline, and 
qualities of leadership developed in Ice- 
land were invaluable in providing cadres 
of experienced Marines around which to 
form these new units. As a result, the 
6th Regiment, which sailed from San 
Diego for New Zealand in late October 
1942, contained only a very small percent- 
age of "Iceland Marines." The military 
wealth had been shared. 


The Marine Corps on the Eve of War ' 


While war came to Europe in Septem- 
ber 1939, the United States did not for- 
mally enter the struggle against the Axis 
Powers for another 27 months. The 
formal declarations of war did not, how- 
ever, project the nation directly from a 
state of isolation and indifference into 
active belligerency. Although the United 
States declared its neutrality — our aim 
being to avoid conflict while guarding 
against totalitarian penetration of the 
Western Hemisphere — we were gradually 
drawn deeper and deeper into short-of- 
war operations in support of Great 
Britain and her allies. 

Initially, the Administration moved 
with caution. In the years following the 
"war to end all wars," disappointment in 
the League of Nation's failure and the 
world-wide depression of the 1930's had 
served to increase our isolationist tend- 
encies. Aware of the national sentiment,^ 
President Eoosevelt initiated a program 
for gradually increasing the armed serv- 
ices, strengthening our bases, and develop- 

' Unless otherwise noted the material in this 
chapter is derived from CMC AnRepts, 1939- 

' The Roper Poll in September 1939 showed 
that extreme interventionist sentiment was 
limited to 2.5% of the total population ; 37% 
preferred to have nothing to do with the warring 
nations. R. E. Sherwood, Roosevelt and Hop- 
kins, An Intimate History (New York: Harper 
and Brother, 1948), 128. 

ing a foundation for the expansion of our 
national resources and industry. On 8 
September 1939, seven days after Hitler's 
armies crossed into Poland, the President 
officially declared a limited national emer- 
gency. As the rising tide of Nazi aggres- 
sion swept over Europe in 1940 and 1941, 
Americans awakened more and more to 
the peril and supported increasingly the 
national policy of strengthening our 
armed forces. 

As of 30 June 1939, two months before 
Hitler's armies launched their Blitzkrieg, 
Marine Corps strength stood at 19,432 of- 
ficers and enlisted,* of whom 4,840 (in- 
cluding aviation components) were as- 
signed to the Fleet Marine Force. FMF 
ground forces were organized in two units 
optimistically designated "brigades," each 
in actuality an understrength infantry 
regiment * reinforced by skeletonized sup- 
porting elements: 1st Brigade based on 
the east coast (Quantico), 2d Brigade 
on the west coast (San Diego). Each 
brigade had the support of a Marine air- 
craft group of corresponding numerical 
designation, and FMF aviation further 
boasted a scouting squadron (VMS-3) 
based in the Virgin Islands. 

However, conversion of international 
tension into armed conflict in Europe re- 
sulted in a marked quickening of United 

'Table DGB-220O-DJF prepared by PersAcct 
Sect, RecordsBr, PersDept, HQMC, 26Nov54. 

*5th and 6th MarRegts of WWI fame, based 
on the east and west coasts respectively. 




States defense efforts. And from that 
point on the Commandant's Annual Re- 
ports reflect a steady succession of upward 
revisions in personnel planning until by 
30 November 1941 total strength stood at 
65,881, the number, give or take a few, 
with which the Marine Corps would enter 
the war against the Axis Powers a week 
later at Pearl Harbor. 

But of greater significance than the 
increase in over-all strength was the grow- 
ing proportion of that strength repre- 
sented by the Fleet Marine Force. Fiscal 
1940 saw the numbers of the Corps' strik- 
ing arm more than doubled : from 4,525 to 
9,749; and this figure in turn had more 
than tripled by 30 November 1941, reach- 
ing 29,532. One factor largely responsible 
for this impressive increase was mobiliza- 
tion in November 1940 of the entire Or- 
ganized Marine Corps Reserve, both 
ground and air, thus making available a 
large number ^ of officers and men, at least 
partially trained, for incorporation into 
the FMF with a minimum of delay. 

This increased strength made possible 
organization of a unit larger than the 
Marine Corps had ever operated before: 
the triangular division, consisting of three 
infantry regiments, an artillery regiment, 
supported by engineer, reconnaissance, 
and signal units plus medical and other 
service troops. Thus on 1 February 1941 
the brigades stationed on the east coast 
and west coast were officially activated 
as the 1st Marine Division and 2d Marine 
Division respectively. To effect the nec- 
essary expansion, cadres were drawn from 
existing units around which to build and 
train new units of the same type. This 
proved a slow and laborious process, and 

months passed before either division could 
be built up to authorized strength. 

Growth of Marine Aviation kept pace 
with that of the ground forces, and again 
that pace looked faster on paper than it 
was in actuality. Simultaneously with 
the conversion of the two brigades into 
divisions, the east coast and west coast 
FMF aircraft groups, based at Quantico 
and San Diego respectively, were acti- 
vated as the 1st and 2d Marine Aircraft 
Wings (MAW). But, as with the divi- 
sions, bringing them up to authorized 
strength proved no overnight process. 

Initially, each could boast only a single 
aircraft group of mixed composition, 
designated MAG-11 and MAG-21 respec- 
tively. On the eve of Pearl Harbor, FMF 
air personnel numbered 2,716 officers and 
enlisted out of a total aviation strength 
of 5,911.* These were divided among the 
two wings and the detached squadron in 
the Virgin Islands. The 1st MAW had 
remained based at Quantico. But the 
coming of war found the 2d MAW scat- 
tered far and wide, with a squadron at 
Wake Island, a detachment at Midway 
Island, and the balance of the wing at 
Ewa, on Oahu, T. H.' 

Though the two divisions and two 
wings comprised the Marine Corps' prin- 
cipal striking arm, considerations of im- 

' Total of 5,241 officers and enlisted. 

" The two groups were identical in composi- 
tion but slightly unequal in strength. Each 
contained 2 fighter, 2 scout-bomber, 1 observa- 
tion, and 1 utility squadrons. MAG-11 had 100 
operational aircraft to 90 for MAG-21. The 
Virgin Islands detachment operated 8 utility- 
scouting planes, bringing the total of FMF air- 
craft of all types to 198. Altogether, Marine 
Aviation included 13 squadrons and 204 opera- 
tional planes of all types. 

' "Administrative History of U. S. Marine 
Corps in World War II" (MS in HistBr Ar- 
chives), 158, hereinafter cited as AdminHist. 



AIR EVACUATION OF WOUNDED /row Ocotal, Nicaragua was pioneered by Marine aviators who 
presaged the mass evacuation techniques of World War II. (USMC 5173) 

ARMY LIGHT TANK is unloaded from its landing craft during joint Army-Marine amphibious exercises 
at New River, N. C. in August I94I. (SC 125129) 

mediate urgency diverted many FMF per- 
sonnel into other activities. The United 
States had no intention of defending 
America on its own soil as long as the 
situation permitted any other choice. The 
Navy already possessed several outlying 

bases and hoped to obtain more, for secu- 
rity of which it relied on the Marines. 
Hence there evolved a type of organiza- 
tion specially adapted to this duty: the 
Marine defense battalion, which was pri- 
marily an artillery outfit whose main 



armament consisted of antiaircraft and 
coast defense guns.® The first four of 
these, with consecutive numerical desig- 
nations, were activated during fiscal 1940. 
By the time of Pearl Harbor the number 
had reached seven with two more in proc- 
ess of formation.^ 

Concurrent with increased numbers 
came increased responsibilities. The 
Navy, too, was expanding at an unprece- 
dented rate, diverting more Marines from 
the FMF to perform the Corps' tradi- 
tional functions : security of naval instal- 
lations ashore and service afloat. By 30 
November 1941, ships' detachments had 
grown to 68, manned by a total of 3,793 

Ashore the Navy's stepped-up training 
programs, particularly in naval aviation, 
created more and more bases, security of 
which imposed a serious additional drain 
on Marine man power. In fiscal 1940 the 
Corps was called upon to provide guard 
detachments at four new naval air sta- 
tions in thfe Continental United States and 
three in U. S. overseas territories." The 
following fiscal year added another four 

* The genesis of the defense battalion was at- 
tributable to two factors: (a) the acceptance 
of the advanced base concept and its logical tac- 
tical requirements; and (b) during the pre- 
World War II period, while the nation was 
apathetic towards rearmament and/or military 
expansion, an increase in Marine strength, un- 
der the guise of a defense force, was politically 
more acceptable. 

°CNO Itr to CMC, 9Dec41, End (a). 

" Compilation from muster rolls closed 
30NOV41 (located at Unit Diary Sect, HQMC). 

*' NAS Key West and Jacksonville, Fla. ; 
Tongue Point, Oreg. ; Alameda, Calif. ; Sitka and 
Kodiak, Alaska ; and San Juan, Puerto Rico. 

air stations, a naval ammunition depot, 
a naval supply depot,^^ and 18 other new 
installations ranging in character and lo- 
cation from David Taylor Basin, Car- 
derock, Maryland, to Naval Magazine, 
Indian Island, Washington. Further- 
more, garrison detachments were 'detailed 
to twelve stations overseas, as will be dis- 
cussed subsequently. 

Simultaneously with filling the Navy's 
demands, the Marine Corps assumed ad- 
ditional security problems of its own as 
existing bases expanded and new ones 
were established. (See below.) Thus, 
the period under discussion saw the acti- 
vation of seven new guard companies of 
a non-FMF character: at Quantico, San 
Diego, Dunedin (Florida), and Bremer- 
ton (Washington). 


Inevitably the problems of housing, 
training, and equipping rapidly expand- 
ing manpower imposed increasing pres- 
sure on the Corps' existing facilities, 
pegged as these were to peacetime needs 
and the economy of depression years. 

Following World War I, activities 
strictly Marine Corps in nature had been 
concentrated generally at the recruit train- 
ing depots at Parris Island and San 
Diego," and at the operational bases at 
Quantico and San Diego, where the East 

" NAS Cape May, N. J. ; Miami, Fla. ; Corpus 
Christi, Tex.; and Quonset Point, R. I.; NAD 
Burns City, Ind. ; NSD, Oakland, Calif. 

" Generally, all recruits from east of the Mis- 
sissippi were trained at Parris Island, all from 
west at San Diego. 



Coast and West Coast components of the 
Fleet Marine Force were stationed. FMF 
aviation was based nearby at MCAS, 
Quantico, and NAS, San Diego. 

Marines first laid eyes on Parris Island 
early in the Civil War when they partici- 
pated in the naval expedition which 
seized adjoining Port Eoyal. This served 
as an important naval base throughout 
the war, but the Navy did not begin con- 
struction of installations on the island 
proper until 1883. The first record of a 
separate Marine detachment setting up 
there permanently occurs in June 1893. 
The post did not begin functioning, how- 
ever, in its present capacity until Novem- 
ber 1915 when the East Coast Marine re- 
cruit depots were transferred there from 
Norfolk and Philadelphia. 

Retained as a permanent base after 
World War I, Parris Island continued its 
role as the point of initial contact with 
military life for all newly enlisted Ma- 
rines from the East. Partly for this rea- 
son, its facilities were maintained at a 
fairly high level during the lean years of 
the 1920's and 1930's. Nevertheless, the 
flood of recruits soon overflowed existing 
facilities and forced a rapid expansion. 
Thus in 1940^1, even as the full train- 
ing program continued and was intensi- 
fied, new barracks, a new post exchange, 
and a new rifle range were added to those 
already operating at full capacity. 

The Recruit Depot, San Diego, which 
had operated as such since August 1923, 
experienced similar problems and arrived 
at similar solutions. As events proved, 
both of these bases managed to keep 
abreast of the expansion program 

throughout the war and thus accomplish 
their basic missions. 

Much of San Diego's success in its pri- 
mary mission was owed to the activation 
of nearby Camp Elliott in mid-1940 to 
furnish advanced training and serve as a 
base for West Coast elements of the FMF. 
Until then San Diego had housed both 
of those activities, and with the speed- 
ing-up expansion program they were be- 
ginning to get in each other's way. The 
first FMF units began the transfer early 
in 1941 and greatly eased the pressure; 
though, as will be seen. Camp Elliott 
itself was eventually pressured out of 

Quantico, acquired by the Marine Corps 
immediately following U. S. entry into 
World War I, found its difficulties less 
readily resolved. During the interim be- 
tween wars, this post assumed a position 
of paramount importance in the develop- 
ment of Marine amphibious doctrine and 
techniques, and in the training of Marine 
officers and technicians. The passage of 
years saw additional educational units 
move in until the Virginia base became 
the center of higher learning for the Ma- 
rine Corps. 

Advent of the national emergency soon 
made it apparent that no practicable phys- 
ical expansion would enable Quantico to 
continue these activities, all rapidly grow- 
ing and intensifying in scope, and at the 
same time serve as home base for east 
coast FMF units, especially when opera- 
tional forces were to reach division size. 
Parris Island, hard pressed to keep 
abreast of its own problems, could do 



little to relieve the presure. Clearly the 
situation called for construction of an en- 
tirely new and extensive base for FMF 
operations on the eastern seaboard. This 
required Congressional approval, which 
was obtained on 15 February 1941. 

The site selected lay in the New River- 
Neuse River area of the North Carolina 
coast. The surveying and purchasing of 
land began immediately. By the end of 
April this preliminary work had been 
completed, and construction of Tent 
Camp #1, Marine Barracks, New River 
commenced. The isolated location of the 
area made development an enormous task. 
Transportation to the site was almost 
nonexistent, electric power lines were 
either lacking or greatly overloaded and 
able to provide but a fraction of the cur- 
rent needed. And the necessary labor 
could be obtained only by offering special 
inducements to workers. Both the Marine 
Corps and civilian contractors approached 
these problems to such good effect that by 
the summer of 1941 the far-from-com- 
pleted camp had reached a stage of devel- 
opment that made it available for use. 

The fledgling 1st Marine Division, still 
understrength,^* moved in shortly after its 
return from maneuvers in the Caribbean. 
There it participated in a series of am- 
phibious exercises, one with the Army's 
1st Infantry Division, the first of four 
Army divisions to receive such training 
jointly with Marine units or under the 
direction of Marine officers. 

Men of the Marine division pitched in 
to improve camp conditions while continu- 
ing their intensive training for combat. 

" While the table of organization listed three, 
the 1st MarDiv had only two infantry regiments 
at this time. 

Civilian contractors pushed construction 
of permanent buildings so effectively that 
soon various specialized training and 
schooling facilities and other units began 
transferring to the new base from both 
Quantico and Parris Island. The 1st Ma- 
rine Division, however, had long since 
departed beyond the seas by the time 
Marine Barracks, New River, reached the 
stage of development where the powers 
that be saw fit to dignify it, late in 1942, 
with the name Camp Lejeune. 

Like the division, the 1st Marine Air- 
craft Wing began outgrowing its Quan- 
tico facilities long before it achieved full 
strength. Even while development pro- 
gressed at New River, the Marine Corps 
obtained authorization for a new air 
base nearby. Cunningham Field, Cherry 
Point, North Carolina, was designated a 
Marine Corps Air Station for develop- 
ment purposes on 1 December 1941, and 
work began on what would become by 
commissioning day, 20 May 1942, a vast 
new base capable of handling the greater 
part of a completely built-up Marine air- 
craft wing.'^ 

On the west coast. Camp Elliott, less 
hampered than Quantico by a multiplicity 
of activities, proved capable initially of 
handling the vastly increased load of ad- 
vanced training, though the camp was ex- 
panded and developed to many times its 
original size in the process. Its 29,000 
acres housed the 2d Marine Division from 
its activation until its departure for the 
Pacific. It also became the home of the 
Marine Corps' first tank training center 
and the infantry training center for 
numerous replacement drafts. 

AdminHist, 159. 




During the years between wars, the per- 
vasive spirit of pacifism which led to re- 
peated attempts by this country to cooper- 
ate in reduction of naval armaments and 
in international treaties militated against 
adequate defense preparations, as did 
budgetary restrictions. Such peace as 
these measures achieved proved uneasy at 
best, but the fact that the U. S. lived up 
to its agreements, whereas some other na- 
tions did not, contributed toward making 
our defense program a shadow of what 
it might have been. This was particu- 
larly serious in the Pacific, as will be seen. 
But in 1939-41, with war flaming through 
Europe, the more immediate danger lay in 
the Atlantic where Hitler's submarines 
appeared nearly invincible. 

In the fall of 1939 the United States 
armed forces were barely adequate for the 
defense of the Western Hemisphere. As 
long as the national sentiment did not 
sanction total i-earmament and military 
expansion, the administration was forced 
to rely on existent means and a partial 
mobilization of both manpower and mate- 
rial. Unfortunately, the lull in military 
operations in Europe during the winter 
of 1939-1940 seemed to justify public 
apathy and made the problem of rearma- 
ment more difficult for the President and 
his military planners. 

Britain's historical dominance of the 
Atlantic sea lanes had given us a false 
sense of security there, and permitted the 
United States to commit a major part of 

"Unless otherwise noted the material in this 
and the following sections is derived from 
Hemisphere Defense; Undeclared War; Strategic 
Planning; Battle of the Atlantic. 

the Navy to guard against Japanese ag- 
gression in the Pacific. However, the 
German offensive in the spring of 1940 
served to jolt Americans from their com- 
placency. German troops overran Den- 
mark, Norway, the Low Countries, and 
France. President Roosevelt recognized 
the danger in this and caused a shift in 
our military policy to provide greater 
security in the Atlantic. 

During the summer and fall of 1940, 
Congress stepped up the procurement of 
aircraft, mobilized the reserves, passed 
selective service legislation, and launched 
the two-ocean navy building program. 
But completion of these measures would 
take time, and we had no assurance that 
the Axis partners would sit idly by and 
enjoy the fruits of their initial aggression. 
To implement the rearmament program, 
President Roosevelt adopted the policy 
of aiding Britain and Russia (after June 
1941) while continuing diplomatic rela- 
tions with Germany and Japan. With 
industry expanding and the armed forces 
increasing in size and equipment, the Ad- 
ministration did everything short of war 
to bolster Britain's tottering position. 

In the fall of 1940 Britain and the 
United States completed negotiations 
which culminated in one of the most ex- 
traordinary military deals in history. 
Britain, holding numerous Caribbean pos- 
sessions, desperately needed additional 
convoy vessels to protect her vital Atlantic 
supply line against submarine depreda- 
tions; the U. S., possessor of numerous 
overage destroyers, wished to strengthen 
defense of eastern approaches to the main- 
land and the Panama Canal. As a result 
of this situation, on 2 September 1940 the 
U. S. agreed to swap 50 of these destroy- 



ers^^ in return for 99-year leases on cer- 
tain base sites in various strategically 
placed British possessions : the Bahamas, 
Jamaica, Antigua, Saint Lucia, Trinidad, 
and British Guiana. 

Since plans called for development of 
these sites into naval activities of varying 
nature, the first Americans to move in 
were Marines of the several security guard 
detachments. The same held true in the 
case of two additional bases not included 
in the destroyer deal: at Argentia (New- 
foundland) and in Bermuda. Thus, while 
in the throes of expanding the FMF, the 
Marine Corps found itself saddled with 
still more garrison duty beyond the con- 
tinental limits of the United States. 


The fall of France and the Netherlands 
alarmed the United States to the danger 
that New World possessions of these 
countries ^^ might fall into Germany's 
hands should Hitler force the conquered 
nations to cede them, or to provide servic- 
ing there for German U-Boats operating 
in the Atlantic. 

Martinique, the administrative and eco- 
nomic center of France's colonies in the 
Caribbean, became the focal point of 

"These were 1,200-toii, flush-deck four-stack- 
ers, vintage of World War I, many of which 
had been laid up since that struggle and had 
to be recommissioned. By the time the trade 
was completed (10Apr41), the U. S. had become 
more deeply involved and threw in an additional 
10 escort vessels of the "Lake" class Coast 
Guard cutter. 

" Dutch Guiana, Aruba, and Curacao ; French 
Guiana on the South American continent; Saint 
Pierre and Miquelon off Newfoundland ; Mar- 
tinique, Guadeloupe, and several smaller islands 
in the West Indies. 

American interest and concern. For 
should the three French warships there, 
including the aircraft carrier Beam 
(loaded with 106 American-manufactured 
fighter planes destined for pre- Vichy 
France) , be taken over by the enemy, the 
security of British and American ship- 
ping in the Atlantic would be seriously 
threatened. Furthermore, the, French 
High Commissioner for the Antilles, Rear 
Admiral Georges Robert had declared his 
allegiance to the Vichy government and 
was emphatic in his refusal to accept 
American and British offers of "protec- 

One solution, and one which was imme- 
diately discarded, called for an American 
break with Vichy and the occupation of 
the islands by American forces. It was 
not expected, however, that Admiral Rob- 
ert would yield without a fight — and we 
were not ready to scrap our neutral pol- 
icy and draw accusations of Yankee im- 
perialism from friendly Western Hemis- 
phere nations. Dire necessity, however, 
required some plan of operation. On 
8 July 1940, the Joint Planning Commit- 
tee completed a plan for an expeditionary 
force, to be readied for embarkation from 
New York on or about 15 July. The 1st 
Marine Brigade ^' was earmarked for the 
initial landing force, to be followed by a 
task force based on the Army's 1st Infan- 
try Division. 

While the expeditionary force was 
readied, officials of the Departments of 
State and Navy worked out a compromise 
to relieve the tense situation. The Ameri- 

" The 1st MarBrig then based at Guantanamo 
Bay, Cuba, was composed of 5th Mar, 1st Bn, 
10th Mar (artillery), 1st EngBn, 1st MedBn, 
three provisional (casual) companies, BrigHqCo, 
and one company each of service troops, chemi- 
cal troops, tanks, and motor transport. 



can representative in negotiations that 
followed, Rear Admiral John W. Green- 
slade, arrived at an agreement with Ad- 
miral Robert to maintain the status quo; 
and the "hot" Martinique problem was 
temporarily resolved without the United 
States being forced into military action. 

However, heightened tensions during 
the late summer of 1940 again indicated 
the possibility of French connivance with 
Germany. Accordingly, late in October 
1940 the President "... asked the Navy 
to draft a plan for an emergency opera- 
tion. . . ." ^° This plan called for an as- 
sault on Martinique, by a naval force in- 
cluding a landing party of some 2,800 
Marines of the 1st Marine Brigade, to be 
supported by two reinforced Army regi- 
ments. Later plans increased the size of 
the force ; revised estimates were based on 
the possibility of more than token resist- 
ance from the seven to eight thousand 
French soldiers and sailors on the island. 

Fortunately, the operation against 
Martinique died stillborn. Admiral 
Greenslade reached a new "gentlemen's 
agreement" with Admiral Robert, al- 
though there were frequent instances later 
when President Roosevelt still thought it 
might be necessary to occupy the island. 
The Marine Corps remained prepared 
for possible action until Admiral Robert 
surrendered his command to American 
Vice Admiral John S. Hoover in June 


As early as spring 1940, President 
Roosevelt was deeply concerned over the 
possibility of a German invasion of the 
Portuguese Azores. These islands lie 
athwart the vital shipping lanes between 

^ Quoted in Hemisphere Defense, Chap IV, 6. 

the United States and the Mediterranean, 
and Europe and South America. While 
the Army considered them of little value 
in Western Hemisphere defense considera- 
tions, their danger was measurable by 
their value to Germany. From air bases 
and naval facilities in the islands, Ger- 
man aircraft and submarines could sortie 
after the bulk of British shipping. 

Our deep concern for the safety and in- 
tegrity of the islands led to a series of dis- 
cussions with both the British, Portugal's 
ally, and the Lisbon government. By 
October 1940, United States Army and 
Navy planning officei"S had drafted a plan 
for a surprise seizure of the Azores. 
However, the plan to land one reinforced 
division was built on sand: the Army did 
not have the necessary troops to commit, 
nor did the Navy have adequate ships to 
transport and support the landing force. 
And, politically, it was contrary to Amer- 
ican policy at this time to become a de 
facto participant in the European war. 

By May 1941 intelligence estimates 
from Europe again indicated the possibil- 
ity of a German movement into the Iber- 
ian peninsula and German occupation of 
the Azores and adjacent islands. On the 
22d of that month, President Roosevelt 
directed the Army and Navy to draft a 
new plan for an expedition to occupy the 
Azores. This plan (GRAY), approved 
by the Joint Board on 29 May, provided 
for a landing force of 28,000 combat 
troops, half Marine and half Army; the 
Navy was responsible for transporting 
and supporting the force. Major General 
H. M. Smith, USMC, would command the 
landing force, under Rear Admiral 
Ernest J. King, the expeditionary com- 

However, while these preparations were 
being made, other factors developed and 



altered the original mission of the mixed 
force. Portugal was opposed to an Amer- 
ican occupation of the Azores, and United 
States planners became preoccupied with 
the threat of German efforts to occupy 
South America, particularly Brazil. The 
succeeding weeks witnessed a change in 
both the urgency for the Azores operation 
and in the mission of the Marine comple- 
ment of the Azores force. 

During the early part of June, intelli- 
gence sources in Europe produced credit- 
able evidence that Germany did not plan 
to invade Spain and Portugal but in- 
tended rather to attack in the opj)osite 
direction. Russia would be Hitler's next 
objective. The forecast of the German 
plans put an end to American fears for 
the safety of the Azores, and permitted 
the United States to divert the Marines 
to Iceland. 


How thin the Marine Corps had to 
spread its manpower in order to fulfill 
its many commitments is indicated by the 
table that follows showing the distribu- 
tion effective 30 November 1941, on the 
eve of Pearl Harbor. The fact that the 
figures quoted do not add up to total 
Corps strength is accounted for by omis- 
sion of minor categories involving indi- 
viduals or small groups of men. 

Continental U. S. (non-FMF) 

Major Marine Corps Bases ^ 14, 707 

Posts & Stations (43) 10,089 

Headquarters & Staff 780 

Recruiting (4 districts) 847 

Total 26, 423 

Overseas (non-FMF) 

Posts & Stations (24) 3,367 

Tactical Units'' 5,498 

Shipboard Detachments (68) 3,793 

Total 12, 658 

Fleet Marine Force, Continental U. S. 

1st MarDiv ___^ 8, 918 

2d MarDiv (less dets) 7,540 

2d DefBn 865 

1st MAW 1,301 

2d MAW (less dets) 682 

Miscellaneous 633 

Total 19, 939 

Fleet Marine Force, Overseas 

5DefBns (Pa:ciflc) 4,399 

2d MAW (elements) (Pacific) 733 

2d MarDiv (elements) (Pacific) 489 

Total 5, 621 

Total above categories 64, 641 

Total strength Marine Corps 65, 881 

^ Quantico, Parris Island, San Diego, Camp 
Elliott, New River. 

"4th Mar (Philippines), 801; 1st SepBn 
(Philippines), 725; 1st MarBrig(Prov) (Ice- 
land), 3,972. 


War Comes 


Prewar Situation in the Pacific 


In the late years of the 19th Century and 
the early decades of the 20th, Japan set 
out to gain more territory. Consistently 
following a policy of encroachment in 
Asia and the Pacific, and retreating only 
when confronted with the threat of supe- 
rior force, the Japanese Empire steadily 
grew in size and strength. Warning sig- 
nals of an impending clash between Japan 
and the Western nations with extensive 
interests in the Orient became increasingly 
evident. In the lOSO's when these nations 
were gripped by economic depression and 
their military expenditures were cut to 
the bone, Japan struck brazenly. 

In 1931 Japanese troops invaded Man- 
churia and no concerted international mil- 
itary effort was made to halt the seizure. 
An ineffectual censure by the I^eague of 
Nations, far from discouraging Japan, 
emboldened her to further action. An- 
grily, the Japanese delegates stalked out 
at Geneva and gave formal notice of in- 
tention to withdraw from the Ijeague. 
The country thickened its curtain of se- 
crecy which shrouded the I^ague-man- 

* Unless otherwise noted the material in this 
section is derived from Senate Doc No. 244, 79th 
Congress, 2d Session, Report of the Joint Com- 
mittee on the Investigation of the Pearl Harbor 
Attack (Washington: GPO, 1946), hereinafter 
cited as Pearl Harbor Rept and the Committee's 
record of 39 volumes of hearings and exhibits, 
hereinafter cited as Hearings Record; G. N. 
Steiger, A History of the Far East (Boston: 
Ginn and Co., 1944). 

dated islands awarded Japan as its share 
of the spoil of German possessions lost in 
World War I. In 1934 the Japanese 
served notice that they would no longer 
abide by the limitations of the Washing- 
ton Naval Disarmament Treaty of 1922. 
Finally, in 1937, Japan attacked China 
and horrified the world with the excesses 
committed by her soldiers in the infamous 
Rape of Nanking. But still there was no 
effective military action to curb this ramp- 
ant aggression. 

In this period Japan was not without 
supporters. Germany and Italy, bent on 
similar programs of territorial aggran- 
dizement in Europe and Africa, made 
common cause with the Japanese. These 
"Axis'' powers signed a mutual assistance 
pact in 1937, ostensibly aimed at the Com- 
munist Cominform, but in essence as a 
show of strength to forestall interference 
with their plans of conquest. In August 
1940, after the outbreak of war in Europe 
and the fall of France, Germany forced 
the Vichy Government to consent to Jap- 
anese occupation of northern Indo-China. 
The three predatory nations combined 
again in less than a month, this time in 
the Tripartite Treaty of 27 September 
which promised concerted action by the 
Axis in case of war with the United States. 

The United States, traditionally a 
friend of China and a supporter of an 
"Open Door'' policy in Asia, strongly op- 
posed Japanese moves to establish hegem- 
ony over the strife-torn Chinese Republic. 




While the political sentiment of the ma- 
jority of Americans in the late 1930's 
would condone no direct military inter- 
vention, the government and the nation 
were openly sympathetic to the Chinese 
cause. Both moral and legal embargoes 
against munitions shipments to Japan 
were put into effect and increasing 
amounts of material aid given to China. 
American pilots, including members of 
the armed forces, were permitted to volun- 
teer to fly for the Chinese Air Force 
against the Japanese.^ 

By early 1941 Japan was hurt in pride, 
purse, and potency as a result of American 
political and economic measures taken to 
halt its expansion. In March a new Am- 
bassador, Admiral Nomura, was sent to 
Washington to negotiate a settlement of 
Japanese- American differences. He was 
confronted with a statement of four prin- 
ciples which represented the basic Ameri- 
can position in negotiations. These were : 

(1) Respect for the territorial integrity and 
the sovereignty of each and all nations ; 

(2) Support of the principle of noninterfer- 
ence in the internal affairs of other countries ; 

(3) Support of the principle of equality, in- 
cluding equality of commercial opportunity ; 

(4) Nondisturbance of the status quo in the 
Pacific except as the status quo may be altered 
by peaceful means.^ 

In retrospect, it seems obvious that 
there was little likelihood of Japan ac- 
cepting any of these principles as a basis 
for negotiations. At the time, however, 
considerable and protracted effort was 
made to resolve differences. Postwar evi- 
dence indicates that the Japanese Premier, 
Prince Konoye, as well as Ambassador 
Nomura were sincere in their efforts to 

achieve a peaceful solution of the threat- 
ening situation in the Pacific. It was 
not Konoye, however, who called the turn 
in Imperial policy, but the Japanese 
Army. And the Army adamantly re- 
fused to consider any concession that 
might cause it to lose face. 

After Germany attacked Russia in June 
1941, the longtime threat of Soviet inter- 
vention in Japan's plans for expansion 
was virtually eliminated. The Japanese 
Army moved swiftly to grab more terri- 
tory and to add to its strength. Southern 
Indo-China was occupied and conscripts 
and reservists were called up. In the face 
of this fresh evidence of Japanese in- 
transigence. President Roosevelt froze all 
Japanese assets in the United States, ef- 
fectively severing the last commercial 
contact between the two nations. 

In October the Army forced the Konoye 
Cabinet to resign and replaced it with a 
government entirely sympathetic to its 
position.* The new premier. General 
To jo, sent a special representative, Saburu 
Kurusu, to Washington to assist Nomura 
and revitalize negotiations. The Japa- 
nese diplomats were in an untenable posi- 
tion. They were instructed, in effect, to 
get the United States to accept Japanese 
territorial seizures on Japanese terms. 
Their mission was hopeless, but behind its 
facade of seeming interest in true negotia- 
tions, Tojo's government speeded up its 
preparations for war. As far as the 
Japanese leaders were concerned, war with 
tlie United States was a now or never 
proposition, since American-inspired eco- 
nomic sanctions would soon rob them of 
the necessary raw materials, particularly 

' United States Relations with China (Wash- 
ington : Dept of State, 1949), 24. 
'Hearings Record, Part 2, 1103-1104. 

*M. Kato, The Lost War (New York: A. A. 
Knopf, 1946), 48. 



oil, which they had to have to supply their 
military machine. 

The only event that might have halted 
Japanese war preparations would have 
been a complete abnegation by the United 
States of its principles of negotiation. 
On 22 November Ambassador Kurusu re- 
ceived the third and last of a series of 
communiques from Japan setting dead- 
lines for successful negotiations. He was 
informed that after 29 November things 
were "automatically going to happen." " 

As far as the Japanese were concerned 
negotiations were at an end and the time 
for direct action had come. The two 
Japanese envoys were carefully instructed, 
however, not to give the impression that 
talks had been broken off. The stage had 
been set for "the day that will live in 

After an extremely thorough investiga- 
tion of the negotiations during this period 
prior to the outbreak of the war, a Joint 
Congressional Committee summed up the 
duplicity of Japanese negotiations in this 
succinct statement : 

In considering the negotiations in their en- 
tirety the conclusion is inescapable that Japan 
had no concessions to make and that her pro- 
gram of aggression was immutable.' 


Both the United States and Japan had 
developed plans for war in the Pacific 
long before December 1941. Each nation 

^ Hearings Rec(yrd, Part 12, Exhibit No. 1, 165. 

"Pearl Harbor Rept, 49. 

' Unless otherwise noted the material in this 
section is derived from Pearl Harbor Rept; 
Hearings Record, Part 13, Exhibits 8-8D, Japan- 
ese Records; USSBS (Pac), NavAnalysisDiv, 
Campaigns of the Pacific War (Washington: 
GPO, 1946), hereinafter cited as Campaigns of 
the Pacific War. 

considered the other to be its most proba- 
ble enemy. There was, however, a funda- 
mental moral difference between the re- 
spective war plans. The Americans 
planned for defense and retaliation in case 
of attack ; the Japanese intended to strike 
the first blow. (See Map 1, Map Section) 

Japan's prime objective was economic 
self-sufficiency, and the prize she sought 
was control of the rich natural resources 
of Southeast Asia and the islands of the 
East Indies, her "Southern Resources 
Area." The Japanese were well aware 
that invasion in this area would bring 
them into conflict with a coalition of 
powers. The lands they aspired to con- 
quer were the possessions or protectorates 
of Great Britain, Australia, New Zealand, 
the Netherlands, and the United States. 
By means of surprise attacks, launched 
simultaneously on a half dozen different 
fronts, the Japanese expected to catch the 
Allies off-balance and ill-prepared. 

The obvious threat of war with Japan 
had not been ignored by any of these Al- 
lied nations, but the tremendous advan- 
tage of choice of time and place of at- 
tack rested with the aggressor. Japan in- 
tended to strike during a period when 
most of the resources in men and material 
of the British Commonwealth were being 
devoted to the defeat of the European 
Axis partners. The Netherlands, which 
existed only as a government-in-exile, 
could contribute quite a few ships but only 
a small number of men to a common de- 
fense force. And the United States, most 
certainly Japan's strongest enemy, was 
heavily committed to support the Allies 
in Europe and the Near East. Moreover, 
that nation was only partially mobilized 
for war. 

The initial Japanese war concept did 
not envisage the occupation of any terri- 



tory east of Tarawa in the Gilberts. All 
operations beyond the limits of the South- 
ern Resources Area were designed to es- 
tablish and protect a defensive perimeter. 
The cordon of strategic bases and island 
outposts was to stretch from the Kuriles 
through Wake Atoll to the Marshalls and 
Gilberts and thence west to the Bismarck 
Archipelago. The islands of Timor, Java, 
and Sumatra in the East Indies were to 
be seized and Japanese troops were to oc- 
cupy the Malayan Peninsula and Burma. 

The major force which might prevent 
or delay the accomplishment of the Japa- 
nese plan was the United States Pacific 
Fleet based at Pearl Harbor. Recogniz- 
ing the threat posed by the American 
naval strength, the Commander in Chief 
of the Japanese Combined Fleet, Admiral 
Isoroku Yamamoto, directed that a study 
be made of the feasibility of a surprise 
aerial attack on Pearl, timed to coincide 
with the outbreak of war. In February 
1941 the first staff considerations of the 
projected raid were begun, but the actual 
details of the operation were not worked 
out until September when it seemed in- 
creasingly obvious to the Japanese high 
command that war was inevitable and that 
they needed this bold stroke to insure the 
success of initial attacks. 

On 3 November the Chief of the Naval 
General Staff, Admiral Osami Nagano, 
approved the draft plan, and on the 5th 
commanders of fleets and task forces were 
given their assignments. Orders were is- 
sued to selected task force units to begin 
moving singly and in small groups to 
Hitokappu Bay in the Kuriles on or about 
15 November. Ten days later a striking 
force, its core six large fleet carriers trans- 
porting the pick of the Japanese Navy's 
planes and pilots, sortied from the se- 
cluded anchorage bound for the Hawaiian 

Islands. The approach route lay well 
north of the search areas patrolled by 
American planes based at Midway and 
Wake and out of normal shipping lanes. 
The tentative day of attack, X-day, had 
been set for a Sunday, 7 December (Pearl 
Harbor time) . Japanese intelligence in- 
dicated that most of the Pacific Fleet 
would be in port on a weekend. Tallies 
of the ships present at the Pearl Harbor 
Naval Base received from the Japanese 
consulate at Honolulu were transmitted 
to the attack force as late as 5 December. 
Vice Admiral Chuichi Nagumo, the strik- 
ing force commander, received orders 
from Yamamoto on 2 December confirm- 
ing the chosen date. There was still time 
to turn back; if the approaching ships 
had been discovered prior to 6 December 
they had orders to return. No one saw 
them, however, and the carriers arrived 
at their launching point right on schedule. 
At midnight of 6-7 December, the Jap- 
anese Combined Fleet Operation Order 
No. 1 informed its readers that a state of 
war existed with the United States, Great 
Britain, and the Netherlands. 


A nation's war plans are never static. 
The constantly changing world political 
scene demands continual reevaluation and 
amendment. In the 1930's, American war 

' Unless otherwise noted the material in this 
section is derived from Pearl Harbor Rept; 
Navy Basic War Plan — Rainbow No. 5 (WPL- 
46), 26May41 and Appendix I, Joint Army-Navy 
Basic War Plan Rainbow No. 5, quoted in full 
in Hearings Record, Part 33, Exhibit No. 4; 
MarCorps Plan C-2, Rainbow No. 5, 5Jun41, 
Plans & Policies Div Files ; M. S. Watson, Chief 
of Staff: Prewar Plans and Preparations — 
United States Army in World War II (Wash- 
ington: HistDiv, DA, 1950); S. E. Morison, 



plans were concerned primarily with 
courses of action to be taken in the event 
of a conflict in one theatre and against 
one nation or a contiguous group of na- 
tions. In these so-called "color plans," 
each probable enemy was assigned a sepa- 
rate color designation; Japan became 
Orange. With the advent of the Axis 
coalition, American military men began 
thinking in terms of a true world war. 
As these new plans evolved they were 
given the name Rainbow to signify their 
concept of a multi-national war. 

The United States was deeply involved 
in the war in Europe soon after its out- 
break, if not as an active belligerent, then 
as the arsenal of the democracies. By 
the spring of 1941 American naval ves- 
sels were convoying shipments of war ma- 
teriel at least part of the way to Europe 
and they were actively guarding against 
German submarines a Neutrality Zone 
that extended far out into the Atlantic. 
The intent of these measui'es and others 
similar to them was clearly to support 
Britain in its war against Germany, Italy, 
and their satellites. There was little ques- 
tion where the syinpathies of the majority 
of Americans lay in this struggle and none 
at all regarding the position of their 

On 29 January 1941, ranking British 
and American staff officers met in Wash- 
ington to discuss joint measures to be 
taken if the United States should be 
forced to a war with the Axis Powers. 
It was regarded as almost certain that 
the outbreak of hostilities with any one 

The Rising Sun in the Pacific 1931-Aprill9},2— 
History of United States Nat'al Operations in 
M^orld War II (Boston: Little, Brown and Com- 
pany, 1948), hereinafter cited as Rising Sun in 
the Pacific. 

of the Axis partners would bring imme- 
diate declarations of war from the others. 
By insuring action on two widely sepa- 
rated fronts, the Axis could expect at the 
very least a decreased Allied capability to 
concentrate their forces. The American- 
British conversations ended on 27 March 
with an agreement (ABC-1) which was 
to have a profound effect on the course of 
World War II. Its basic strategical de- 
cision, which never was discarded, stated 
that : 

Since Germany is ttie predominant member of 
the Axis Powers, the Atlantic and European 
area is considered to be the decisive theatre. 
The principal United States military effort will 
be exerted in that theatre, and operations of 
United States forces in other theatres will be 
conducted in such a manner as to facilitate that 
effort .... If Japan does enter the war, the 
Military Strategy in the Far East will be 

The defensive implied in the war 
against Japan was not to be a holding 
action, however, but rather a strategic de- 
fensive that contemplated a series of tac- 
tical offensives with the Pacific Fleet as 
the striking force. A new American war 
plan. Rainbow 5, was promulgated soon 
after the end of the American-British 
talks. Almost the whole of the Pacific 
was made an American strategic responsi- 
bility and the Army's primary mission 
under the plan was cooperation with and 
support of the fleet. 

A listing of the contemplated offensive 
actions of Rainbow 5, which included the 
capture of the Caroline and Marshall Is- 
lands, would be interesting but academic. 
The success of the Japanese raid on Pearl 
Harbor forced a drastic revision of strat- 
egy which effectively postponed amphib- 

" Para 13 of ABC-1 quoted in Hearings Rec- 
ord, Part 33, 958. 



ious assaults in the Central Pacific. Cer- 
tain defensive measures which were men- 
tioned in the plan, however, were imple- 
mented prior to the outbreak of war and 
in most of them Marine forces figured 

Some of the Marine defense battalions, 
tailored to meet the needs of garrisons for 
isolated island outposts, were already in 
the Pacific by the time Rainbow 5 was 
published. The plan called for the de- 
velopment of bases, primarily air bases, at 
Midway, Johnston, Palmyra, Samoa, and 
Wake. All of these islands, which were 
under control of the Navy, were to have 
Marine garrisons. Guam, in the center of 
the Japanese-held Marianas, which had 
long had a small Marine barracks detach- 
ment, w^as decisively written off in the war 
plan; its early capture by the Japanese 
was conceded. The rest of the islands 
were placed in a category which called for 
defense forces sufficient to repel major 

The purpose of establishing bases on 
these islands was twofold. Samoa was to 
help protect the routes of communication 
to the Southwest Pacific; Johnston, Pal- 
myra, Wake, and Midway were to serve as 
outguards for the Pacific Fleet's home 
port at Pearl. (See Map 1, Map Section) 


The Navy did not start cold with its ad- 
vance base development scheme for the 
four island outposts of the Hawaiian 

" Unless otherwise noted the material in this 
section is derived from CNO Serial 070412, 
23Jun41, "Policy regarding employment of Mar- 
ine Defense Battalions in the Pacific Area" 
(located at NRMC) ; CXO Serial 091812, 
25Sep41, "Employment of Marine Defense Bat- 
talions" ; CO, 1st DefBn Itr to OIC, HistDiv, 

Group. A blueprint for base expansion 
in the Pacific had been laid out in the re- 
port of the Navy's Hepburn Board, a 
Congressionally authorized fact-finding 
group which, in the spring of 1938, made 
a strategic study of the need for addi- 
tional United States naval bases. The 
potential utility of Midway, Wake, John- 
ston, and Palmyra was recognized," and 
surveys were conducted and plans made 
for the construction of base facilities, air- 
fields, and seadromes during 1939 and 
1940. The responsibility for developing 
garrison plans and locating coastal and 
antiaircraft gun positions Avas given to 
Colonel Harry K. Pickett, 14th Naval Dis- 
trict Marine Officer and Commanding 
Officer, Marine Barracks, Pearl Harbor 
Navy Yard. The fact that Colonel Pickett 
personally surveyed most of the base sites 
insured active and knowledgeable coopera- 
tion at Pearl Harbor with requests from 
the islands for men and materiel to imple- 
ment the garrison plans. 

Although they were popularly referred 
to in the singular sense, a custom that will 
be continued in this narrative, each of the 
outposts was actually a coral atoll encom- 
passing varying numbers of bleak, low- 

HQMC, 29Dec43; CO, 3d DefBn Itr to OIC, 
HistDiv, HQMC, 4Feb44 ; MD, 1st DefBn, 
Palmyrals, Annual Rept of Activities, lJul43; 
Hist of the 7th DefBn, 21Dec42 ; 1st SamoanBn, 
MCR, Annual Rept of Activities, lJul42 ; LtCol 
R. D. Heinl, The Defence of Wake (Washington : 
HistSec, PublnfoDiv, HQMC, 1947), hereinafter 
cited as Defense of Wake; LtCol R. D. Heinl, 
Marines at Midivay (Washington: HistSec, 
PublnfoDiv, HQMC, 1&48), hereinafter cited as 
Marines at Midicay. 

" House Doc No. 65, 76th Congress, 1st Ses- 
sion, "Report on the Need of Additional Naval 
Bases to Defend the Coast of the United States, 
its Territories and Possessions" (Hepburn 
Board Rept), 3Jan39, passim. 



lying sand islands within a fringing reef. 
Each atoll had at least one island big 
enough to contain an airstrip; Midway 
had two. The lagoons within the reefs 
were all large enough to permit the dredg- 
ing and blasting of seaplane landing 
lanes and anchorages for small cargo 
ships; Midway's and Wake's were also 
slated for development as forward bases 
for the Pacific Fleet's submarines. Civil- 
ian contractors were hired to build the na- 
val base installations, but until war actu- 
ally broke out most of the work on the 
island defenses was done by the men who 
were to man them, Marines of the 1st, 3d, 
and 6th Defense Battalions. 

The organization of the defense battal- 
ions varied according to time and place 
of employment, but by late 1941 the stand- 
ard T/0 called for a unit with more than 
900 men assigned to a headquarters bat- 
tery, three 5-inch coast defense gun bat- 
teries, three 3-inch antiaircraft batteries, 
a sound locator and searchlight battery, a 
battery of .50 caliber antiaircraft machine 
guns, and a battery' of .30 caliber machine 
guns for beach defense. Midway was the 
only outpost that actually drew an entire 
battalion, although Wake originally was 
slated to be garrisoned by one. On John- 
ston and Palmyra the habitable area was 
so limited that it was impossible to accom- 
modate more than a small defense detach- 

Some development work had been done 
on Wake and Midway, the two northern 
islands, before the arrival of the naval 
contractors' construction crews. In 1935 
Pan American World Airways had set up 
way stations for its Clipper service to the 
Orient on both Midway and Wake and a 
relay station of the trans-Pacific cable had 
been in operation on Midway's Sand Is- 

land since 1903. Most construction, like 
the passenger hotel on Wake and the quar- 
ters for the airline's and cable company's 
personnel, was of little military value. 

Midway, which had the most ambitious 
base plan, was also the first outpost sched- 
uled to receive a Marine garrison — the 3d 
Defense Battalion which arrived at Pearl 
Harbor on 7 May 1940. The bulk of the 
battalion remained in Hawaii for the next 
eight months while reconnaissance details, 
followed by small advance parties, did 
the preliminary work on supply and de- 
fense installations.^^ On 27 January 
1941, in the face of the threat posed by 
Japan's aggressive actions, the Chief of 
Naval Operations (CNO) directed that 
the rest of the 3d Defense Battalion be 
moved to Midway, that detachments of 
the 1st Defense Battalion be established 
at Johnston and Palmyra, and that the 
6th Defense Battalion, then in training at 
San Diego, move to Pearl Harbor as a re- 
placement and reserve unit for the out- 

On 15 February, the same day that the 
3d Battalion began unloading its heavy 
equipment at Midway, an advance detach- 
ment of the 1st Defense Battalion left San 
Diego on the Enterprise. At Pearl Har- 
bor the detachment left the carrier and 
transferred to a small cargo ship that 
steamed on to the southwest for 800 miles 

"BriGen A. R. Pefley notes on draft manu- 
script, 14Jan57. Since all fresh water had to be 
distilled, the capacity of the distillers set the 
limit for the size of the island garrison. In terms 
of water consumption each contractor's work- 
man took the place of a Marine. Adm C. C. 
Bloch Itr to ACofS, G-3, HQMC, 7Jan57. 

"CNO Serial 0618, 17Jan41, "Establishment 
of Permanent Marine Defense Forces at John- 
ston, Midway, and Palmyra Islands." 



to reach tiny Johnston where on 3 March 
two 5-inch guns, six Marines, and two 
naval corpsmen were set ashore. After a 
few days layover to help the caretaker de- 
tail get set up, the rest of the advance 
party (3 officers and 45 enlisted men) 
went on to Palmyra, approximately 1,100 
miles south of Oahu. 

After the remainder of the 1st Defense 
Battalion arrived at Pearl, small reinforc- 
ing detachments were gradually added to 
the southern outpost garrisons as the is- 
lands' supply and quartering facilities 
Avere expanded. On Johnston and Pal- 
myra, as at Midway, the civilian contrac- 
tors' crews and construction equipment 
were heavily committed to the naval air 
base program, and only occasionally 
could the Marines borrow a bulldozer, 
truck, or grader to help out in their own 
extensive schedule of defense construction. 
For the most part, the garrisons relied on 
pick and shovel to get their guns emplaced 
and to dig in the ammunition magazines, 
command posts, and fire direction centers 
necessary for island defense. 

Duty on the small atolls was arduous 
and dull with little relief fi'om the monot- 
ony of a steady round of work and train- 
ing. When a few hours off was granted, 
there was no place to go and little to do; 
the visible world shrank to a few uninvit- 
ing acres of dunes, scrub brush, and coral 
surrounded by seemingly endless stretches 
of ocean. The visits of patrol planes, sup- 
ply ships, and even inspection parties 
were welcomed. Under the circumstances, 
morale at the isolated posts remained sur- 
prisingly high, helped perhaps by the 
prospect of action. 

In so far as possible, the 14th Naval 
District attempted to follow a policy of 

rotation for the men at the outlying posts, 
replacing those that had been longest "in 
the field" with men from Pearl Harbor. 
In midsummer a group of 1st Defense 
Battalion personnel was sent to Midway 
to start the relief of the 3d Battalion and 
on 11 September the 6th Defense Battal- 
ion arrived to take over as the atoll's gar- 
rison. The 3d Battalion returned to 
Hawaii for a well-deserved break from 
the gruelling monotony and work of 
building defenses. 

By August 1941 the work on the naval 
air base at Wake was well along and the 
need for a garrison there was imperative. 
An advance detachment of the 1st De- 
fense Battalion arrived at the atoll on 
19 August and immediately began the 
now familiar process of backbreaking 
work to dig in guns, dumps, aid stations, 
and command posts. Again the contrac- 
tor's men and machines were largely de- 
voted to work on the airfield and the 
lagoon, and the Marines had to get along 
with the hand tools organic to the unit. 
In late October reinforcments from the 
parent battalion made the 2,000-mile trip 
from Hawaii to bring the garrison up to 
a strength of nearly 400 men. The unit 
scheduled to be the permanent garrison 
on Wake, the 4th Defense Battalion, ar- 
rived at Pearl Harbor on 1 December, too 
late to reinforce or replace the Wake De- 
tachment. A most important addition to 
the atoll's defenses did arrive, however, 
before war broke. Twelve Grumman 
Wildcats of Marine Fighter Squadron 
211 flew in to the airstrip off the Enter- 
prise on 4 December. 

Just before the Japanese attacked, the 
strength of defense battalion personnel on 
outpost duty and at Pearl Harbor was: 



Pearl Harbor 















1st DefBn _____ _._ . 













3d DefBn _ _._ 


4th DefBn .._ 

6th DefBn 



For armament the outposts relied 
mainly on the organic weapons of the de- 
fense battalions: 5-inch naval guns, 
3-inch antiaircraft guns, and .30 and .50 
caliber machine guns. Midway had, in 
addition, three 7-inch naval guns still to 
be mounted and a fourth gun at Pearl 
Harbor waiting to be shipped. The 
breakdown of weapon strength showed : ^* 





5-inch guns 





3-inch guns 





.50cal MGs__ 





.30cal MGs__ 





Although the list of weapons was impos- 
ing, the garrisons were not strong enough 
to man them adequately ; the standard de- 
fense battalion of 1941, moreover, in- 
cluded no infantry. 

In contrast to the garrisons of the Pearl 
Harbor outposts, the 7th Defense Battal- 
ion slated for duty at Tutuila, main island 
of American Samoa, was a composite in- 
fantry-artillery unit. The battalion Avas 
organized at San Diego on 16 December 
1940 with an initial strength of 25 officers 
and 392 enlisted men. Its T/0 called for 

■* ComFourteen Rept of Status of DefBns as- 
signed to the 14th ND, lDec41 (located at 
NRMC). Personnel figures include naval medi- 
cal personnel assigned to the defense battalions. 

a headquarters company, an infantry com- 
pany, and an artillery battery as well as a 
small detail which had the mission of or- 
ganizing and training a battalion of 
Samoan reservists. 

The islands of American Samoa had a 
native population of almost 10,000 which 
could be drawn upon as a labor force and 
for troops to back up a regular garrison. 
This was not the only significant differ- 
ence between the outpost atolls and Samoa, 
however. The terrain of Tutuila, which 
was by far the largest and most heavily 
populated of the islands, was mountain- 
ous and heavily forested, and its 52 square 
miles contained a number of areas that 
could be converted into camps and supply 
depots. There was room for training 
areas and small arms ranges. The fine 
harbor at Pago Pago, site of the U. S. 
Naval Station and headquarters of the 
naval governor, could be used by large 
vessels. This combination of harbor, el- 
bow room, and an indigenous labor force, 
plus its location along the shipping route 
to the Southwest Pacific, inade Tutuila 
a vital strategic base. (See Map 3) 

During the spring and early summer 
of 1940, Major Alfred R. Pefley of 
Colonel Pickett's staff made a thorough 
survey of Tutuila and prepared a detailed 
plan for its defense. On 29 November the 
CNO directed that defense plans based on 
Pefley's recommendations be implemented 



immediately. The naval governor was au- 
thorized to begin construction of coast de- 
fense and antiaircraft gun positions. 
Most of the guns to be mounted were al- 
ready in storage at the naval station and 
the Bureau of Ordnance was directed to 
provide the ammunition and additional 
weapons still needed.^^ 

The primary purpose of raising the 7th 
Defense Battalion was the manning of the 
four 6-inch naval guns and six 3-inch 
antiaircraft guns provided for in initial 
defense plans. The wisdom of including 
infantry in the battalion and making pro- 
vision for reinforcement by trained Sa- 
moan reserves can hardly be questioned. 
Tutuila was far too large an island to be 
adequately protected by a relatively few 
big guns, most of which were concen- 
trated around Pago Pago harbor. Small 
beach defense garrisons were needed all 
around the island shorelines to check 
enemy raiding parties. It was intended 
that most of the Samoan reserves would 
be equipped and trained with rifles taken 
from naval stores and used in the beach 
defenses where their knowledge of the 
terrain would be invaluable. 

An advance party of the 7th Defense 
Battalion, which left the States before the 
unit was formally activated, arrived at 
Pago Pago on 21 December 1940. The 
rest of the battalion made the 4,500-mile 
voyage from San Diego via Pearl Harbor 
in March, arriving on the 15th. The next 
months were busy ones as guns were em- 
placed and test fired, beach defenses were 
constructed, miles of communication lines 
were laid, and trails were cut Avhich would 
enable quick reinforcement of threatened 
landing points. 

It was midsummer before the first Sa- 
moan Marine was actually enlisted, but 
many natives voluntarily took weapons 
training on an unpaid status, continuing 
a practice begun by the naval governor in 
November 1940.^*^ The first native recruit 
was enlisted on 16 August 1941 and the 
1st Samoan Battalion, Marine Corps Re- 
serve, was a going concern by the time 
war broke. The authorized strength of 
the battalion was 500 enlisted men, but 
this figure could never be reached because 
of the great number of men needed as 
laborers on essential base construction. 

There was one factor of the defense 
picture at Tutuila that matched the situ- 
ation at Midway, Johnston, and Palmyra. 
None of these islands had, at the onset 
of war, any land planes. The Marine air 
squadrons which were scheduled to join 
the defenders were either still in the 
States or else based on Oahu, waiting for 
the signal that the airfields were ready 
for use. That part of Marine Air which 
was in the Hawaiian Islands was based at 
Ewa Field, located approximately four 
air miles west of Pearl Harbor. Just 
prior to the Japanese attack, the units 
stationed at the field were Headquarters 
and Service Squadron of Marine Aircraft 
Group 21 (MAG-21) ; Marine Scout 
Bomber Squadron 232 (VMSB-232) ; 
Marine Utility Squadron 252 (VMJ- 
252) ; and the rear echelon of VMF-211, 
which had moved forward to Wake. Op- 
erational control of the Marine planes in 
the Hawaiian area was exercised by the 
Commander Aircraft, Battle Force, Pa- 
cific Fleet." 

"CNO serial 054430, 29Nov40, "Defense of 
American Samoa." 

"Gov of AmerSamoa Itr to CNO, 13Feb41, 
"Establishment of Native Insular Force." 

" 2dLt B. Hollingshead, "The Japanese Attack 
on 7 December 1941 on the Marine Corps Air 



Aside from the Marine forces in the 
Western Pacific assigned to the Asiatic 
Fleet," the only sizeable Marine units in 
the Pacific not already accounted for were 
guard detachments on Oahu and the 2d 
Engineer Battalion (less Companies C 
and D) which had been sent to Oahu to 
establish an advance amphibious training 
base for the 2d Marine Division. There 
was a 485-man Marine Barracks at the 
Pearl Harbor Navy Yard and 102 men 
assigned to the barracks at the Naval Air 

Station at Ewa, Oahu, Territory of Hawaii" 
(MS, HistDiv, HQMC, January 1945), 3-8, here- 
inafter cited as Ewa Monoijraph. The other 
squadrons assigned to MAG-21 were either at 
sea with the Navy's carriers or still in the U. S. 
" See Part IV, "Marines in the Philippines," 
for the prewar situation in China and the Philip- 

Station at Ford Island. Marines pro- 
vided the guard (169 men) at the Naval 
Ammunition Depot at Lualualei in the 
hills northwest of Honolulu. The defense 
battalions which were quartered in or near 
the navy yard were under the operational 
control of the Commanding Officer, Ma- 
rine Barracks, Colonel Pickett. 

There were an additional 877 Marines 
present in Pearl Harbor on 7 December 
as members of the guard detachments of 
the battleships and cruisers of the Pacific 
Fleet." In all, there were more than 
4,500 Marines on Oahu (hat first day. 

" The strength of most Marine units on Oahu 
is listed in Hearings Record, Part 24, Exhibit 
No. 40. "Location of regularly assigned com- 
manding officers of ships present during the 
Japanese attack of 7 December 1941." 

Japan Strikes 



Perhaps no action in American mili- 
tary history has been so thoroughly docu- 
mented, examined, and dissected as the 
Pearl Harbor attack. Investigation has 
followed investigation ; a host of books 
have been written on the subject, all in an 
effort to pin down the responsibility in 
the welter of charge and countercharge. 
The issue of what individuals or set of 
circumstances, if any, should bear the 
blame for the success of the Japanese raid 
has not been, and may never be, finally 
decided. On one point, however, there 
has been unanimous agreement — that the 
courage of the vast majority of defending 
troops was of a high order. 

The first inkling of the Japanese attack 
came not from the air, but from the sea. 
At 0637 on 7 December, more than an 
hour before any enemy planes were 
sighted, an American patrol bomber and 
the destroyer Ward attacked and sank an 
unidentified submarine in the restricted 
waters close to the entrance to Pearl Har- 
bor.^ This vessel was one of five Japa- 

* Unless otherwise noted the material in this 
section is derived from Pearl Harbor Rept; 
Hearings Record, Part 13, Exhibits Nos. 8-8D, 
Japanese Records, and Parts 23 and 24, Hear- 
ings and Exhibits of the Roberts Commission ; 
MarFor, 14th ND Jnl, December 1941 ; Ewa 
Monograph; Col H. K. Pickett Itr to BriGen 
C. D. Barrett, 22Dec41 (located at NRMC, Job 
6608, Box 25) ; Rising Sun in the Paeific. 

' "Unfortunately, the radio report sent to the 
14th N. D. was not clear, and in view of many 

nese two-man submarines which had the 
extremely risky mission of penetrating the 
Pacific Fleet's stronghold. The midgets 
were transported to the target on board 
large long-range submarines, part of an 
undersea scouting and screening force 
which had fanned out ahead of the enemy 
carriers. Not one of the midget raiders 
achieved any success ; four were sunk and 
one ran aground. 

The Japanese attack schedule allowed 
the Americans little time to evaluate the 
significance of the submarine sighting. 
The first enemy strike group was airborne 
and winging its way toward Oahu be- 
fore the Ward fired its initial spread of 
depth charges. The Japanese carrier 
force had turned in the night and steamed 
full ahead for its target, launching the 
first plane at 0600 when the ships were ap- 
proximately 200 miles north of Pearl 
Harbor. A second strike group took off at 
0745 when the carriers had reached a posi- 
tion 30 miles closer to the American base. 
Although a radar set on the island picked 
up the approaching planes in time to give 
warning, the report of the sighting was 
believed an error and disregarded, and 
the Japanese fighters and bombers ap- 
peared unannounced over their objectives. 

The enemy plan of attack was simple. 
Dive bombers and fighter planes would 

previous false reports, it was considered neces- 
sary to check the report. The air attack started 
before verification was received." Adm C. C. 
Block Itr, op. cit. 




strafe and bomb the major Army and 
Navy airfields in an attempt to catch de- 
fending aircraft on the ground. Simul- 
taneously, the battleships moored to pil- 
ings along the shore of Ford Island 
would be hit by high- and low-level bomb- 
ing attacks. The shipping strike groups 
included large numbers of dive and hori- 
zontal bombers, since the Japanese antici- 
pated that protective netting might pre- 
vent their lethal torpedo bombers from 
being fully effective. In all, 321 planes 
took part in the raid, while 39 fighters 
flew protective cover over the carriers to 
guard against a retaliatory attack that 
never materialized. 

At 0755 the soft stillness of Sunday 
morning was broken by the screaming 
whine of dive bombers and the sharp chat- 
ter of machine guns. At half a dozen 
different bases around the island of Oahu 
Japanese planes signaled the outbreak of 
war with a torrent of sudden death. Pa- 
trol bombers were caught in the water at 
Kaneohe Naval Air Station, across the 
island from Honolulu; closely parked 
rows of planes, concentrated to protect 
them from sabotage, w'ere transformed 
into smoking heaps of useless w^reckage at 
the Army's Wlieeler and Hickam Fields, 
the Marines' air base at Ewa, and the 
Navy's Ford Island air station. The at- 
tack on the airfields had barely started 
before the first bombs and toi-pedoes were 
loosed against the sitting targets of "battle- 
ship row." Within minutes most of the 
battleships at the Ford Island moorings 
had been hit by one or more torpedoes 
and bombs. If the Japanese had drawn 
off after the first fifteen minutes of their 
attacks, the damage done would have been 
terrific, but the enemy planes kept on 
strafing and bombing and the toll of ships, 
planes, and men soared. 

The Americans did not take their beat- 
ing lying down. The first scattered shots 
from sentries ashore and watch standers 
who manned antiaircraft guns on board 
ship flashed back at the enemy even be- 
fore the bugles and boatswains' pipes 
sounded "Call to Arms" and "General 
Quarters." The ships of the Pacific Fleet 
were on partial alert even in port and 
most of the officers and men were on 
board. Crew members poured up the lad- 
ders and passages from their berthing 
compartments to battle stations. While 
damage control teams tried to put down 
fires and shore up weakened bulkheads, 
gun crews let loose everything they had 
against the oncoming planes. In many 
cases guns were fired from positions 
awash as ships settled to the bottom and 
crewmen were seared with flames from 
fuel and ammunition fires as they con- 
tinued to serve their weapons even after 
receiving orders to abandon ship. On 
many vessels the first torpedoes and bombs 
trapped men below deck and snuffed out 
the lives of others before they were even 
aware that the attack was on. 

The reaction to the Japanese raid was 
fully as rapid at shore bases as it was on 
board ship, but the men at the airfields 
and the navy yard had far less to fight 
with. There was no ready ammunition at 
any antiaircraft gun position on the is- 
land; muzzles impotently pointed sky- 
ward while trucks were hurried to muni- 
tions depots. Small arms were broken 
out of armories at every point under at- 
tack; individuals manned the machine 
guns of damaged aircraft. The rage to 
strike back at the Japanese was so strong 
that men even fired pistols at the enemy 
planes as they swooped low to strafe. 

At Ewa every Marine plane was 
knocked out of action in the first attack. 


JAPANESE PHOTOGRAPH of the sneak attack at Pearl Harbor showing the line of American battle- 
ships caught at their mooring near Ford Island. ( USN 30550 ) 

JAPANESE LANDING ON GUAM depicted by a propaganda artist who shows a portion of the troops 
and transports of the South Seas Detached Force. ( SC 301 167 ) 

Two squadrons of Japanese fighters swept 
in from the northwest at 1,000 feet and 
dived down to rake the aircraft parked 
near the runways with machine-gun and 
cannon fire. Pilots and air crewmen ran 
to their planes in an attempt to get them 

into the air or drag them out of the line 
of fire, but the Japanese returned again 
and again to complete the job of destruc- 
tion. When the enemy fighters drew off 
at about 0825 they left behind a field lit- 
tered with burning and shot-up aircraft. 



The men of MAG-21 recovered quickly 
from their initial surprise and shock and 
fought back with what few rifles and ma- 
chine guns they had. Salvageable guns 
were stripped from damaged planes and 
set up on hastily improvised mounts; one 
scout-bomber rear machine gun was 
manned to swell the volume of antiair- 
craft fire. Although the group command- 
er, Lieutenant Colonel Claude A. Larkin, 
had been wounded almost as soon as he 
arrived at the field that morning, he con- 
tinued to coordinate the efforts to meet 
further enemy attacks. 

Two Japanese dive bombers streaked 
over the field from the direction of Pearl 
Harbor at 0835, dropping light fragmen- 
tation bombs and strafing the Marine gun 
positions. A few minutes after the 
bombers left, the first of a steady proces- 
sion of enemy fighters attacked Ewa as 
the Japanese began assembling a cover 
force at nearby Barber's Point to protect 
the withdrawl of their strike groups. The 
Marine machine guns accounted for at 
least one of the enemy planes and claimed 
another probable. Two and three plane 
sections of fighters orbited over the field, 
and occasionally dived to strafe the gun- 
ners, until the last elements of the Japa- 
nese attack force headed out to sea around 

Three of the Marine airmen were killed 
during tlie attacks, a fourth died of 
wounds ; 13 wounded men were treated in 
the group's aid station. Flames demol- 
ished 33 of the 47 planes at the field ; all 
but two of the remainder suffered major 
damage. The sole bright note in the pic- 
ture of destruction was the fact that 18 of 
VMSB-231's planes were on board the 
Lexington^ scheduled for a fly-off to Mid- 
way, and thereby saved from the enemy 

Within the same half hour that wit- 
nessed the loss of Ewa's planes, the possi- 
bility of effective aerial resistance was 
canceled out by similar enemy attacks all 
over Oahu. Ford Island's seaplane ramps 
and runways were made a shambles of 
wrecked and burning aircraft in the open- 
ing stage of the Japanese assault. The 
Marines of the air station's guard detach- 
ment manned rifles and machine guns to 
beat off further enemy thrusts, but the 
dive bombers had done their job well. 
There was no need for them to return. The 
focus of all attacks became the larger ships 
in the harbor. 

The raid drew automatic reactions from 
the few Marines in the navy yard who saw 
the first enemy planes diving on the ships. 
While the guard bugler broke the major- 
ity of the men of the barracks detachment 
and the 1st and 3d Defense Battalions out 
of their quarters, the early risers were al- 
ready running for the armories and gun 
sheds. By 0801 when Colonel Pickett or- 
dered the defense battalion machine-gun 
groups to man their weapons, eight of the 
guns had already been set up. More ma- 
chine guns were hastily put in position 
and men were detailed to belt the ammuni- 
tion needed to feed them, while rifle am- 
munition was issued to the hundreds of 
men assembled on the barracks' parade 
ground. Pickett ordered the 3-inch anti- 
aircraft guns in the defense battalions' 
reserve supplies to be taken out of storage 
and emplaced on the parade. He dis- 
patched trucks and working parties of the 
2d Engineer Battalion to Lualualei, 27 
miles up in the hills, to get the necessary 
3-inch shells. The Marine engineers also 



sent their heavy earth-moving equipment 
to Hickam Field to help clear the runways. 

Thirteen machine guns were in action 
by 0820 and the gunners had already ac- 
counted for their first enemy dive bomber. 
During the next hour and a half the fire 
of twenty-five more .30's and .50"s was 
added to the yard's antiaircraft defenses, 
and two more planes, one claimed jointly 
with the ships, were shot down. The 
3-inch guns were never able to get into ac- 
tion. The ammunition trucks did not re- 
turn from the Lualualei depot until 1100, 
more than an hour after the last Japanese 
•aircraft had headed back for their carriers. 
By that time the personnel of all Marine 
organizations in the navy yard area had 
been pooled to reinforce the guard and 
antiaircraft defense, to provide an infan- 
try reserve, and to furnish the supporting 
transport and supply details needed to 
sustain them. 

In the course of their attacks on battle- 
ship row and the ships in the navy yard's 
drydocks, the enemy planes had strafed 
and bombed the Marine barracks area, and 
nine men had been wounded. They were 
cared for in the dressing stations which 
Pickett had ordered set up at the begin- 
ning of the raid to accommodate the flow 
of wounded from the stricken ships in the 
harbor. Many of these casualties were 
members of the Marine ship detachments ; 
102 sea-going Marines had been killed dur- 
ing the raid, six later died of wounds, and 
49 were wounded in action.^ 

The enemy pilots had scored heavily : 
four battleships, one mine layer, and a tar- 

' Casualty figures were compiled from records 
furnished by Statistics Unit, PersAcctSec, 
PersDept, HQMC. 

get ship sunk; four battleships, three 
cruisers, three destroyers, and three auxil- 
iaries damaged. Most of the damaged 
ships required extensive repairs. Ameri- 
can plane losses were equally high : 188 
aircraft totally destroyed and 31 more 
damaged. The Navy and Marine Corps 
had 2,086 officers and men killed, the Army 
194, as a result of the attack; 1,109 men 
of all the services survived their wounds. 

Balanced against the staggering Ameri- 
can totals was a fantastically light tally 
sheet of Japanese losses. The enemy car- 
riers recovered all but 29 of the planes they 
had sent out ; ship losses amounted to five 
midget submarines; and less than a hun- 
dred men were killed. 

Despite extensive search missions floAvn 
from Oahu and from the Enterprise, 
which was less than 175 miles from port 
when the sneak attack occurred, the enemy 
striking force was able to withdraw un- 
detected and unscathed. In one respect 
the Japanese were disappointed with the 
results of their raid; they had hoped to 
catch the Pacific Fleet's carriers berthed 
at Pearl Harbor. Fortunately, the urgent 
need for Marine planes to strengthen the 
outpost defenses had sent the Lexington 
and the Enterprise to sea on aircraft ferry- 
ing missions. The Enterprise was return- 
ing to Pearl on 7 Decem.ber after having 
flown off VMF-211's fighters to Wake, and 
the Lexington, enroute to Midway with 
VMSB-231's planes, turned back when 
news of the attack was received. Had 
either or both of the carriers been sunk 
or damaged at Pearl Harbor, the outlook 
for the first months of the war would have 
been even more dismal. The Japanese suc- 
cess had the effect of delaying the schedule 
of retaliatory attacks and amphibious op- 



erations in the Central Pacific that had 
been outlined in Rainbow 5. A complete 
reevaluation of Pacific strategy was nec- 

The critical situation facing the outpost 
islands was clearly appreciated and an at- 
tempt was made to get reinforcements to 
Wake before the Japanese struck; it did 
not come in time. The tiny atoll was one 
of the first objectives on the enemy time- 
table of conquest.* Midway was more for- 
tunate; when the Lexington returned to 
Pearl on 10 December with its undelivered 
load of Marine scout bombers, they were 
ordered to attempt an over-water flight to 
the atoll. On 17 December, ten days after 
the originally scheduled fly-off, 17 planes 
of VMSB-231, shepherded by a naval pa- 
trol bomber, successfully made the 1,137- 
mile flight from Oahu to Midway. It was 
the longest single-engine landplane massed 
flight on record, but more important it 
marked a vital addition to Midway's de- 
fensive potential. 

The outpost islands needed men and 
materiel as well as planes. Rear Admiral 
Claude C. Bloch, Commandant of the 14th 
Naval District, gave the responsibility for 
organizing and equipping these reinforce- 
ments to Colonel Pickett. On 13 Decem- 
ber, all Marine ground troops in the dis- 
trict were placed under Pickett as Com- 
manding Officer, Marine Forces, 14th 
Naval District. The necessary reinforce- 
ments to be sent to Midway, Johnston, and 
Palmyra were drawn from the 1st, 3d, and 
4th Defense Battalions. By the month's 
end the first substantial increments of men, 
guns, and equipment had been received at 
each of the outposts.^ They were not safe 

* For the detailed story of the defense of Wake 
see Part III. 

^CO, MarFor, 14th ND Itr to MGC, 5Jan42, 
Development of outpost garrisons. 

from attacks by any means, but their posi- 
tions were markedly stronger. 


The Washington Naval Disarmament 
Treaty of 1922 provided for the mainte- 
nance of the statvs quo in regard to forti- 
fications and naval bases in certain areas of 
the Pacific. American adherence to these 
terms through the 14-year life of the 
treaty had the practical effect of weaken- 
ing the defenses of the Philippines and 
preventing the development of Guam as a 
naval stronghold. The Hepburn Board of 
1938 recommended that Guam be heavily 
fortified and garrisoned, ^ but Congress 
failed to authorize the expenditure of the 
necessary funds. Unhappily, the planners 
of Rainbow 5 had to concede the capture 
of the island in the first stages of a war 
with the Japanese. It was almost as if 
they could look over enemy shoulders and 
see the terse direction to the commander 
of the Japanese Fourth Fleet to "invade 
Wake and Guam as quickly as possible" ® 
at the onset of hostilities. (See Map 2) 

Guam was a fueling station for naval 
vessels making the long run to and from 
the Orient, a relay point for the trans- 
Pacific cable, the site of a naval radio sta- 
tion, and a stop for Pan American clip- 
pers. Assigned to protect its 20,000 
natives and its 228 square miles of rugged, 
jungled terrain was a token force of 153 

" Unless otherwise noted the material in this 
section is derived from Maj O. R. Lodge, The Re- 
capture of Oumn (Washington : HistBr, G-3 Div, 
HQMC, 1954), 7-9; Capt G. J. McMillin, USN, 
OflRept to the CNO of the Surrender of Guam 
to the Japanese, llSep45; T. Wilds, "The Japa- 
nese Seizure of Guam," MC Gazette, July 195.5. 

' Hepburn Board Rept, op. cit. 

^Hearings Record, Part 13, Exhibit No. 8-C, 
CombFlt OpOrd No. 1, 5Nov41, 475. 



Marines. Backing them up was a Guam- 
anian infantry unit, the 80-man Insular 
Force Guard, and a volunteer native 
naval militia with 246 ill-armed and ill- 
trained members." The island's govern- 
ment departments and naval station ac- 
tivities were manned by 271 regular Navy 
personnel. A naval officer, Captain 
George J. McMillin, was both island gov- 
ernor and garrison commander. 

The war threat was so real by October 
1941 that all women and children of U. S. 
citizenship were evacuated from Guam. 
On 6 December the garrison destroyed all 
its classified papers and like other Pacific 
outposts awaited the outcome of the U. S.- 
Japanese negotiations in Washington. 
The word came at 0545 on 8 December (7 
December, Pearl Harbor time). Captain 
McMillin was informed of the enemy at- 
tack by the Commander in Chief of the 
Asiatic Fleet. In less than three hours 
Saipan-based Japanese bombers were over 
the island. 

The initial enemy target was the mine 
sweeper USS Penguin in Apra Harbor; 
this small ship's 3-inch and .50 caliber 
guns were the only weapons larger than 
.30 caliber machine guns available to the 
Guam garrison. Under repeated attacks, 
the Penguin went to the bottom, and her 
survivors joined the forces ashore. The 
attack continued throughout the daylight 
hours with flights of bombers hitting the 
various naval installations and strafing 
roads and villages. The island capital, 
Agana, was cleared of civilians, and the 

"The members of the Insular Force Guard 
were in the U. S. Government service and re- 
ceived 50 percent of the pay of corresponding 
ratings in the U. S. Navy. The native militia 
served without pay and had no arms except 
obsolete and condemned rifles. RAdm G. .J. Mc- 
Millin Itr to CMC, 3Nov52. 

few local Japanese were rounded up and 

That night a native dugout landed near 
Ritidian Point on the northern cape of the 
island, and the three men in it were cap- 
tured. They claimed to be Saipan natives 
sent over to be on hand as interpreters 
when the Japanese landed. These natives 
insisted that the Japanese intended to land 
the next morning (9 December) on 
beaches near Agana. Captain McMillin 
suspected a trick. He believed that by 
this ruse the Japanese sought to draw the 
Marines out of their prepared positions in 
the butts of the rifle range at Sumay on 
Orote Peninsula. He decided not to allow 
this information to cause a shift of his 
major defensive force from a position 
which guarded important Apra Harbor. 

By guess or knowledge the Saipan na- 
tives had one of the landing sites located 
accurately, but they were off on their time. 
The 9th brought no landing, but the 
bombers came back to give Guam another 
pounding. The Insular Force Guard was 
posted to protect government buildings in 
Agana, but the rest of the island's garri- 
son remained at their assigned posts. 
Lieutenant Colonel William K. McNulty's 
122 Marines of the Sumay barracks con- 
tinued to improve their rifle range de- 
fenses, and the 28 Marines who were as- 
signed to the Insular Patrol, the island's 
police force, kept their stations in villages 
throughout Guam. 

After the Japanese bombers finished for 
the day all was quiet until about 0400 on 
10 December. At that time flares burst 
over Dungcas Beach north of Agana, and 
some 400 Japanese sailors of the 5th De- 
fense Force from Saipan came ashore. 
While the naval landing party moved into 
Agana where it clashed with the Insular 
Force Guard, elements of the Japanese 





10 DECEMBER 1941 


I 9 ,1 ^ ? " 


MAP 2 



South Seas Detached Force (approxi- 
mately 5,500 men) ^° made separate land- 
ings at Tumon Bay in the north, on the 
southwest coast near Merizo, and on the 
eastern shore of the island at Talafofo 

At Agana's plaza the lightly-armed 
Guamanians, commanded by Marine First 
Lieutenant Charles S. Todd, stood off the 
early Japanese attacks, but their rifles and 
machine guns did not provide enough fire- 
power to hold against a coordinated attack 
by the Dungcas Beacli landing force. 
Captain McMillin, aware of the over- 
whelming superiority of the enemy, de- 
cided not to endanger the lives of the thou- 
sands of civilians in his charge by further 
and fruitless resistance. "The situation 
Avas simply hopeless," he later related." 
He surrendered the island to the Japanese 
naval commander shortly after 0600, and 
sent orders to the Marines at Sumay not 
to resist. The word did not reach all de- 
fenders, however, and scattered fighting 
continued throughout tlie day as the 
enemy spread out to complete occupation 
of the island. But this amounted to only 
token I'esistance. There was no chance 
that the determined Japanese might be 
driven off by a force so small, even if the 
defenders could have regrouped. Guam 
had fallen, and it would be two and a half 
years before the United States Avas in a 
position to win it back. 

" This reinforced brigade, commanded by 
MajGen Tomitara Horii, had been organized in 
November 1941 to take part in the capture of 
Guam and to move on from there to seize Rabaul 
in the Bismarcks. It was built around tlie 144th 
InfRegt and reinforced by units of the .Japanese 
55th Division. MIDiv, WD, Order of Battle for 
Japanese Armed Forces, lMar45, 122. 

" McMillin Surrender Rept, op. cit. 

During the tAvo days of bombing and in 
the fighting on 10 December, the total gar- 
rison losses were 19 killed and 42 wounded 
including four Marines killed and 12 
wounded." The civilian population suf- 
fered comparable but undetermined casu- 
alties. The Japanese evacuated American 
members of the garrison to prison camps 
in Japan on 10 January 1942, and the 
enemy naval force that had been present 
at the surrender settled down to duty as 
occupation troops. 


Part of the Japanese striking force 
which raided Pearl Harbor was a task unit 
of two destroyers and a tanker which pro- 
ceeded independently from Tokyo Bay to 
a separate target — Midway. The mission 
of the destroyers Avas implied in their 
designation as the MidAvay Neutralization 
Unit; they Avere to shell the atoll's air base 
on the night of 7 December Avhile the Jap- 
anese carrier force retired from the Ha- 
Avaiian area. (See Map 10, Map Section) 

DaAvn of 7 December found five sea- 
planes of MidAvay's patrol bomber squad- 
ron (VP-21) aloft on routine search mis- 
sions; tAA'o other (Dutch) patrol bombers 
had just taken off for Wake, next leg of 
their journey to the Netherlands East 
Indies. On the Sand Island seaplane 
ramp tAvo more PBYs (Catalina patrol 

^ Marine casualty figures were compiled from 
records furnished by the Statistics Unit. 
PersAcctSec, PersDept, HQMC. 

" Unless otherwise noted the material in this 
section is derived from Hearings Recoi'ds, Part 
24, Exhibit No. 34, "History of Action Occurring 
at Midway 7-31Dec41 as Compiled from Official 
Dispatches and Correspondence," Exhibit No. 35, 
CO, NAS, Midway Itr to CinCPac, Action of 
7Dec41, n. d., and Exhibit No. 36, CO, 6th DefBn 
Itr to ComFourteen, Rept of action on the night 
of 7Dec41, 12Dec41 ; Marines at Midtcau. 



bombers) were warming up to guide in 
VMSB-231 which was scheduled to fly 
off the Lexington that day. At 0630 
(0900 Pearl Harbor time) a Navy radio 
operator's signal from Oahu flashed the 
first news of the Pearl Harbor attack. A 
few minutes later a dispatch from Ad- 
miral Bloch confirmed this repoi't and di- 
rected that current war plans be placed in 

Commander Cyril T. Simard, the Is- 
land Commander, recalled the Dutch 
PBYs (which were then put to use 
by VP-21), established additional air 
search sectors, and ordered Lieutenant 
Colonel Harold D. Shannon's 6th Defense 
Battalion to general quarters. The re- 
mainder of the day was spent in prepara- 
tion for blackout, and in issuing ammuni- 
tion, digging foxholes, and testing com- 
munications. All lights and navigational 
aids were extinguished after it was learned 
that the Lexington, with VMSB-231 still 
on board, had been diverted to seek the en- 
emy's Pearl Harbor striking force. 

Air searches returned late in the day 
without having sighted any signs of Japa- 
nese ships or planes, and the atoll but- 
toned up for the night with all defensive 
positions fully manned. At 1842, a Ma- 
rine lookout saw a flashing light some dis- 
tance southwest of Sand Island, but it 
quickly disappeared, and it was about 
2130 before the one operational radar on 
Sand began picking up what seemed to be 
surface targets in the same general direc- 
tion. Simultaneously two other observ- 
ers, equipped with powerful 8x56 night 
glasses, reported seeing "shapes" to 

Shannon's searchlight battery com- 
mander. First Lieutenant Alfred L. Booth, 
requested permission to illuminate, but his 
request was turned down. Senior officers 

did not want to risk premature disclosure 
of defensive positions. It was also er- 
roneously believed that friendly ships 
were in the area, and there were strict 
orders against illuminating or firing 
without specific orders." 

The apprehension of these observers 
was justified. The Japanese destroyers 
Akebono and Ushio had left their tanker 
Shifiya at a rendezvous point some 15 
miles away and made landfall on the atoll 
at about 2130. By the time Lieutenant 
Booth had been cautioned about his search- 
lights, the two enemy ships had their guns 
trained on Midway and were ready to 
make their first firing run. The firing 
began at 2135. 

The first salvos fell short, but as the de- 
stroyers closed range on a northeast course 
the shells began to explode on Sand Is- 
land. The initial hits struck near Bat- 
tery A's 5-inch seacoast guns at the south 
end of the island, and subsequent rounds 
bracketed the island's power plant, a re- 
inforced concrete structure used also as the 
command post of a .50 caliber antiaircraft 
machine-gun platoon. One round came 
through an air vent and exploded inside 
the building. The Japanese ships then 
suspended fire while they closed on the 
atoll for a second firing run. 

In the island's power plant First Lieu- 
tenant George H. Cannon, although se- 
verely wounded, directed the re-establish- 
ment of wrecked communications and the 
evacuation of other wounded. He refused 
evacuation for his own wounds until after 
Corporal Harold R. Hazelwood had put 
the switchboard back in operation. Can- 
non died a few minutes after reaching the 

" LtCol A. L. Booth Itr to CMC, 27Jan48, here- 
inafter cited as Booth; LtCol L. S. Fraser Itr to 
CMC, hereinafter cited as Fraser. 



aid station, but for this action lie received 
posthumous award of the Medal of Honor. 
He was the first Marine so honored in 
World War II. 

Meanwhile the enemy ships opened fire 
again, this time at closer range, and Com- 
mander Simard ordered Shannon to en- 
gage targets of opportunity. Japanese 
shells set the roof of the seaplane hangar 
on Sand ablaze, lighting up the target for 
the enemy gunners, and accurate salvos 
struck the Pan American radio installa- 
tion, the island laundry, and adjacent 
shops. At 2153 the Marine searchlight 
crews got Shannon's orders to illuminate, 
but by then only the light on the south end 
of Sand could bear on the ships. This 
light silhouetted the Akehono about 2,500 
yards south of the island, before a near 
miss from one of the destroyers put it out 
of commission. Crewmen reacted imme- 
diately to get the light back in action and 
on target, but Battery A's 5-inchers stayed 
silent because communication damage had 
prevented passing of Shannon's command 
to open fire." 

But Captain Jean H. Buckner, com- 
manding Battery D's 3-inch antiaircraft 
guns, could now see the large Japanese 
battle flag on the Akehono'' s foremast, and 
he ordered his guns into action. Splashes 
could not be made out, although illumina- 
tion was excellent, and Buckner's fire con- 
trolmen were positive that the shells were 
either passing through the ships' super- 
structures or into their hulls. Battery B 
(First Lieutenant Rodney M. Handley) 
on Eastern Island now added its 5-inch 
fire to the battle and .50 caliber machine 
guns opened up on the targets which were 
well within range. This firing from the 

Marine batteries kept up for five minutes 
before the Japanese succeeded in knocking 
out the searchlight. Although some ob- 
servers believed that the Ushio had also 
been hulled, results of this Marine fire 
have never been determined.^* Both Japa- 
nese ships retired soon after the light was 
shot out and a Pan American clipper 
captain flying overhead that night en route 
from Wake reported seeing an intense fire 
on the surface of the sea and the wakes of 
two ships on the logical retirement course 
of the destroyers. Both enemy ships, 
however, returned to Japan safely, despite 
any damage that might have been done by 
the Marine guns. 

The enemy fire had cost the 6th Defense 
Battalion two killed and ten wounded ; ^^ 
two men from the naval air station were 
killed and nine wounded. Material dam- 
age on Midway was not too severe and was 
confined to Sand Island; the airfield on 
Eastern Island was not touched. The sea- 
plane hangar had burned, although the 
frame was still intact, and one plane was 
lost in the flames. Another PBY was 
badly damaged by shell fragments, and 
fragments also caused minor damage to a 
number of buildings. The garrison had 
stood off its first Japanese attack, but 
there was little comfort in this. The de- 
fenders estimated — correctly — that the 
enemy would be back sooner or later with 
a much more serious threat. 

With the outbreak of war, completion 
of the coastal and antiaircraft defenses of 
Midway took first priority and Marines 
were treated to the welcome and unusual 

'''Booth; Fraacr; Col L. A Hohn Utr to CMC, 

" The Ushio, evidently a very lucky ship, was 

the only enemy vessel that took part in the Pearl 

Harbor attack that was still afloat on V-J Day. 

" Casualty figures were compiled from records 

" Casualty figures were compiled from records 

furnished by Statistics Unit, PersAcctSec, 



sight of the civilian contractor's heavy 
equipment turned to on dugout and bat- 
tery construction. Authorities at Pearl 
Harbor were determined to get reinforce- 
ments to the atoll and within a week after 
VMSB-231 made its historic long flight 
from Oahu, two batteries of the 4th De- 
fense Battalion with additional naval 3- 
inch and 7-inch guns for coast defense 
were being unloaded. On Christmas, the 
Brewster Buffaloes of VMF-221 flew in 
from the Saratoga which had been rushed 
out to Pearl from San Diego after the 
Japanese attack. This carrier had taken 
part in the abortive attempt to relieve 
Wake. The next day the island received 
another contingent of 4th Defense Bat- 
talion men, the ground echelon of VMF- 
221, and much needed defense materiel 
when the seaplane tender Tangier, which 
had also been headed for Wake, unloaded 
at Midway instead. By the end of Decem- 
ber the atoll, which was now Hawaii's 
most important outpost, had for its garri- 
son a heavily reinforced defense battalion, 
a Marine scout-bomber and a fighter 
squadron, and VP-2rs patrol bombers. 
Midway was in good shape to greet the 
Japanese if they came back, and the pas- 
sage of every month in the new year 
made the atoll a tougher nut to crack.^* 


Tiny Johnston Island, set off by itself 
in the open sea southwest of Hawaii, 
proved to be a favorite target of Japanese 

" See Part V, "The Battle of Midway" for the 
story of the events leading up to the decisive 
naval action which took place at Midway in June 

'" Unless otherwise noted the material in this 
section is derived from Hearings Record, Part 24, 
Exhibit No. 27, "History of Action Occurring at 
Palmyra Island 7-31Dec41, As Compiled from 

submarines in the first month of the war. 
It was too close to the Pacific Fleet base at 
Pearl and too limited in area to make it a 
prize worth risking an amphibious as- 
sault, but its strategic location, like an ar- 
rowhead pointing at the Japanese Mar- 
shalls, made damage to its air facilities 
well worth the risk of bombardment at- 
tempts. The airfield on the atoll's name- 
sake, Johnston Island, was only partially 
completed on 7 December, but temporary 
seaplane handling facilities were in oper- 
ation at Sand Islet, the only other land 
area within the fringing reef. There was 
no permanent patrol plane complement, 
but Johnston was an important refueling 
stop and a couple of PBYs were usually 
anchored in the lagoon. 

The news of the outbreak of war cre- 
ated a flurry of activity on Johnston, and 
the civilian contractor's employees turned 
to at top speed to erect additional earth- 
works around the Marine guns and to pre- 
pare bomb shelters.^" No Japanese ship or 
submarine made its appearance on 7 De- 
cember, perhaps because the first day of 
war found the Indianapolis and five de- 
stroyer minesweepers at Johnston testing 
the performance of the Higgins landing 
boat on coral reefs.^^ These ships were 

Official Dispatches and Correspondence," Exhibit 
No. 28, "History of the Action Occurring at John- 
ston Island 7-31Dec41, As Compiled from OflS- 
cial Dispatches and Correspondence," Exhibit 
No. 31, CO, NAS, Palmyrals Itr to ComFourteen, 
24Dec41„ and Exhibit No. 32, CO, NAS, John- 
stonls Itr to ComFourteen, 19Dec41 ; CO, NAS, 
Johnstonis Itr to ComFourteen, 22Dec41 ; Mar- 
GarFor, Pac File C-1455-40-5, "Defense-Fortifi- 
cation Johnston Island," 12Sep41-13Jun43 ; Mar- 
GarFor, Pac File C-14.55-40-15, "Defense-Forti- 
flcation Palmyra Islands," 26Sep41-30Jun43. 

'"CO, NAS, Johnstonis, Progress and Read- 
iness Rept, 15Dec41. 

=' Hearings Record, Part 23, 758-759. 



immediately recalled toward Pearl to 
form part of the extensive search pattern 
for the enemy carrier force, and John- 
ston's defense rested with its own slim 
garrison. Major Francis B. Loomis, Jr., 
Executive Officer of the 1st Defense Bat- 
talion, caught while returning to Pearl 
by air from an inspection of the western 
outposts, assumed command of the John- 
ston detachment as senior Marine officer 

Shortly after dark on 12 December a 
submarine surfaced 8,000 yards off Sand 
Islet and began firing green star clusters 
which burst high over the island. The 5- 
inch battery could not pick up the vessel 
in its sights, but it fired one star shell in 
the general direction of the submarine. 
The submarine ceased firing immediately 
as she evidently was not seeking a duel. 

The next enemy attack came at dusk 
three days later. The supply ship Bur- 
rows had delivered a barge load of sup- 
plies originally intended for the Wake 
garrison and picked up 77 civilian con- 
struction employees for return to Pearl 
when a sentry atop Johnston's water 
tower spotted a flash to seaward and 
sounded general quarters. The flash had 
been spotted by the batteries also, and the 
5-inch control estimated the range at 9,000 
yards. The 3-inch director and height 
finder made out two ships, one larger than 
the other. The first two enemy salvos 
bracketed Johnston and the third struck 
near the contractor's power house and set 
off a 1,200-gallon oil tank which imme- 
diately fired the building. A strong wind 
whipped up 50-foot flames from the oil 
fire, and "as observed from the Naval Air 
Station at Sand Islet, Johnston Island 

seemed doomed." ^^ The Japanese con- 
tinued to fire for ten minutes at this well- 
lighted target and they hit several other 
buildings. The 5-inch guns delivered 
searching fire, and just as the Marines 
were convinced they were hitting close 
aboard their targets, the enemy fire ceased 

The enemy vessels had fired from the 
obscuring mists of a small squall and spot- 
ters ashore never clearly saw their targets, 
but the defenders believed that they had 
engaged two surface vessels, probably a 
light cruiser and a destroyer. Later anal- 
ysis indicated, however, that one or more 
submarines had made this attack. For- 
tunately no one in the garrison was hurt 
by the enemy fire, although flames and 
fragments caused considerable damage to 
the power house and water distilling ma- 
chinery. The Burrows, although clearly 
outlined by the fire, was not harmed. The 
fact that its anchorage area was known 
to be studded with submerged coral heads 
probably discouraged the Japanese from 
attempting an underwater attack, and 
Johnston's 5-inch battery ruled out a sur- 
face approach. 

During the exchange of fire one of the 
Marines' 5-inch guns went out of action. 
Its counter-recoil mechanism failed. After 
this the long-range defense of the island 
rested with one gun until 18 December 
when two patrol bombers from Pearl ar- 
rived to join the garrison. This gun was 
enough, however, to scare off an enemy 
submarine which fired star shells over 
Sand Islet after dark on 21 December. 
Again the simple expedient of firing in 
the probable direction of the enemy was 
enough to silence the submarine. The 

"-Jftirf., Part 24, Exhibit No. 32, CO, NAS, 
Johnstonis Itr to ComFourteen, 19Dec41. 


next night, just as the ready duty PBY 
landed in the lagoon, another submarine, 
perhaps the same one that had fired illumi- 
nation over Sand, fired six shells at the 
islets. Both 5-inchers on Johnston now 
were back in action and each gun fired ten 
rounds before the submarine submerged. 
The patrol plane was just lifting from the 
water as the last enemy shot was fired. 
Only one shell hit Sand, but that one 
knocked down the CAA homing tower and 
slightly wounded one Marine. 

Johnston Island was clearly a discour- 
aging place to attack, and the shelling of 
22 December marked the last enemy at- 
tempt at surface bombardment. It was 
just as well that the Japanese decided to 
avoid Johnston, because reinforcement 
from Pearl soon had the atoll bursting at 
its seams with men and guns. An addi- 
tional 5-inch and a 3-inch battery, 16 more 
machine guns, and the men to man them 
arrived on 30 December. In January a 
provisional infantry company was sent 
and eventually the garrison included even 
light tanks. The expected permanent Ma- 
rine fighter complement never got settled 
in at Johnston's airfield. The island be- 
came instead a ferrying and refueling stop 
for planes going between Pearl and the 
South and Southwest Pacific. 

Palmyra, 900 miles southeast of John- 
ston, also figured in the early develop- 
ment of a safe plane route to the southern 
theater of war. But before the atoll faded 
from the action reports it too got a taste 
of the gunfire of a Japanese submarine. 
At dawn on 24 December an enemy raider 

surfaced 3,000 yards south of the main 
island and began firing on the dredge 
Sacramento which was anchored in the 
lagoon and clearly visible between two 
of Palmyra's numerous tiny islets. Only 
one hit was registered before the fire of 
the 5-inch battery drove the submarine 
imder. Damage to the dredge was minor 
and no one was injured. 

Colonel Pickett's command at Pearl 
Harbor had oi'ganized strong reinforce- 
ments for Palmyra and these arrived be- 
fore the end of December. Lieutenant 
Colonel Bert A. Bone, Commanding Offi- 
cer of the 1st Defense Battalion, arrived 
with the additional men, guns, and equip- 
ment to assume command of the defense 
force. On 1 March the official designa- 
tion of the Marine garrison on Palmyra, 
was changed to 1st Defense Battalion and 
former 1st Battalion men at other bases 
were absorbed by local commands. The 
Marine Detachment at Johnston became a 
separate unit. 

After these submarine attacks of De- 
cember, Palmyra and Johnston drop from 
the pages of an operational history. The 
atolls had served their purpose well ; they 
guarded a vulnerable flank of the Hawai- 
ian Islands at a time when such protection 
was a necessity. While the scene of active 
fighting shifted westward the garrisons 
remained alert, and when conditions per- 
mitted it many of the men who had served 
out the first hectic days of the war on these 
lonely specks in the ocean moved on to 
the beachheads of the South and Central 

The Southern Lifeline 



In December 1941 reverse followed re- 
verse in the fortunes of the Allies in the 
Pacific. The Japanese seemed to be every- 
where at once and everywhere successful. 
Setbacks to the enemy schedule of con- 
quest were infrequent and temporary. 
On the Asian mainland Hong Kong fell 
and Japanese troops advanced steadily 
down the Malay Peninsula toward Singa- 
pore. In the Philippines Manila was evac- 
uated and American-Filipino forces re- 
treated to Bataan and Corregidor for a 
last-ditch stand. To the south the first 
Japanese landing had been made on 
Borneo, and superior enemy forces pre- 
pared to seize the Netherlands East 
Indies. The capture of Wake and Guam 
gave the Japanese effective control over 
the Central Pacific from the China coast 
to Midway and Johnston. (See Map 1, 
Map Section) 

By the turn of the year only the sea 
area between the Hawaiian Islands and 
the United States and the supply route 
from the States through the South Pa- 

' Unless otherwise noted the material in this 
section is derived from The War Reports of Gen- 
eral of the Army Oeorge C. Marshall — General of 
the Army H. H. Arnold — Fleet Admiral Ernest 
J. King (Philadelphia & New York: J. B. Lip- 
pincott Company, 1947), hereinafter cited as War 
Repwts; FAdm E. J. King and Cdr W. M. White- 
hill, Fleet Admiral King: A Naval Record (New 
York: W. W. Norton & Company Inc., 1952) 
hereinafter cited as King's Naval Record; Stra- 
tegic Planning. 

cific to New Zealand and Australia were 
still in Allied hands. The responsibility 
for holding open the lines of communica- 
tion to the Anzac area ^ rested primarily 
with the U. S. Pacific Fleet. On 31 De- 
cember that fleet came under the command 
of the man who was to direct its opera- 
tions until Japan unconditionally sur- 
rendered—Admiral Chester W. Nimitz 

As soon as he arrived at Pearl Harbor, 
Nimitz was given a dispatch from Ad- 
miral Ernest J. King, the newly ap- 
pointed Commander in Chief, United 
States Fleet (CinCUS, later abbreviated 
as CominCh). King's message outlined 
Nimitz's two primary tasks as CinCPac. 
He was to use his ships, planes, and men 

( 1 ) Covering and holding the Hawaii-Midway 
line and maintaining its communications with 
the west coast. 

(2) Maintaining communications between the 
west coast and Australia, chiefly by covering, 
securing and holding the Hawaii-Samoa line, 
which should be extended to include Fiji at the 
earliest possible date.' 

Although the Japanese had severely 
damaged the Pacific Fleet in their Pearl 
Harbor raid, they had concentrated on 

^ Anzac is actually the abbreviation for Aus- 
tralian and New Zealand Army Corps used in 
WW I, but the term was so understandable and 
easy to use in reference to the two Common- 
wealth nations that it was adopted in the Pa- 
cific War and applied frequently to the geo- 
graphic area in which they lay. 

^King's Naval Record, 353-354. 




ships rather than installations, and the re- 
pair facilities of the navy yard were vir- 
tually untouched. Round-the-clock work 
promptly restored to operation many ves- 
sels which might otherwise have been lost 
for good or long delayed in their return to 
fleet service. But Nimitz's strength was 
not enough to hazard a large scale am- 
phibious offensive, even with the addition 
of reinforcements sent from the Atlantic 
Fleet. In the first few months of 1942, 
Allied strategists had to be content with 
defensive operations. The few local at- 
tacks they mounted were hit-and-run raids 
which did little more than boost home- 
front and service morale at a time when 
most news dealt with defeat and sur- 

From 22 December to 14 January, the 
political and military leaders of the 
United States and Great Britain met in 
Washington (the ARCADIA Confer- 
ence) to chart the course of Allied opera- 
tions against the Axis powers. The 
Americans, despite the enormity of the 
Japanese attack, reaffirmed their decision 
of ABC-1 that Germany was the pre- 
dominant enemy and its defeat would be 
decisive in the outcome of the war. The 
Pacific was hardly considered a secondary 
theater, but the main strength of the Al- 
lied war effort was to be applied in the 
European, African, and Middle Eastern 
areas. Sufficient men and materiel would 
be committed to the battle against Japan 
to allow the gradual assumption of the 

One result of the ARCADIA meetings 
was the organization of the Combined 
Chiefs of Staff (CCS), a supreme mili- 
tary council whose members were the 
chiefs of services in Great Britain and the 
United States. The CCS was charged 

with the strategic direction of the war, 
subject only to the review of the political 
heads of state. The necessity of present- 
ing a united American view in CCS dis- 
cussions led directly to the formation of 
the United States Joint Chiefs of Staff 
( JCS) as the controlling agency of Amer- 
ican military operations. 

On 9 February 1942, the first for- 
mal meeting of General George C. Mar- 
shall (Chief of Staff, United States 
Army), Lieutenant General Henry H. 
Arnold (Chief of the Army Air Corps), 
Admiral Harold R. Stark (CNO), and 
Admiral King (CominCh) took place. 
fExcept for the combination of the offices 
of CominCh and CNO in the person of 
Admiral King which took effect on 26 
March (Admiral Stark became Com- 
mander U. S. Naval Forces Europe) and 
the addition of Admiral William D. 
Leahy as Chief of Staff to the President 
on 20 July, the membership of the JCS 
remained constant for the duration of the 
war. As far as the Marine Corps was 
concerned their representative on the JCS 
was Admiral King, and he was consist- 
ently a champion of the use of Marines at 
their greatest potential — as specially 
trained and equipped amphibious assault 

* On 13Apr51, before a subcommittee of the 
Senate Committee on Armed Services, Gen Hol- 
comb stated that he was called in during the 
ARCADIA conferences and "sat as a member 
of that group." Later "... a formal organi- 
zation occurred in which I was not included. 
However, because of my intercourse with Ad- 
miral Stark I was in on nearly all of the dis- 
cussions that took place." This intimate rela- 
tionship changed, however, when Stark was re- 
lieved as CNO on 26Mar42. An interesting sequel 
to this story of the "exclusion" of the Comman- 
dant from the JCS was revealed by Gen Holcomb 
when he further related how after a dinner party 



On 10 January 1942, the CCS, acting 
with the approval of Prime Minister 
Churchill and President Roosevelt, set up 
a unified, inter-Allied command in the 
western Pacific to control defensive oj)era- 
tions against the Japanese along a broad 
sweep of positions from Burma through 
Luzon to New Guinea. The commander 
of ABDA (American-British-Dutch -Aus- 
tralian) forces holding the barrier zone 
was the British Commander in Chief in 
India, General Sir Archibald P. Wavell; 
his ABDA air, naval, and ground com- 
manders Avei-e respectively , an English- 
man, an American, and a Dutchman. But 
ABDA Command had no chance to stop 
the Japanese in the East Indies, Malaya, 
or the Philippines. Wavell's forces were 
beaten back, cut off, or defeated before he 
could be reached by reinforcements that 
could make a significant difference in the 
fighting. By the end of February Singa- 
pore had fallen and the ABDA area w^as 
split by an enemy thrust to Sumatra. 
Wavell retui'ned to India to muster troops 
to block Japanese encroachment into 

at the White House in July 1943, the President, 
associating himself with the Marine Corps, had 
said to him confidentially : "You know, the first 
thing you know we are going to be left out of 
things. We are not represented on the Joint 
Chiefs of Staff .... how would you like to be 
a member of the Joint Chiefs of Staff?" Holcomb 
replied that he would like it very much but 
didn't know how the Joint Chiefs would feel 
about it. That was the last, however, that Hol- 
comb ever heard of this matter directly or ofl9- 
cially. Senate Committee on Armed Services, 
82d Congress, Hearings on S. 667, "A Bill to 
Fix the Personnel Strength of the United States 
Marine Corfis and to make the Commandant of 
the Marine Corps a Permanent Member of the 
Joint Chiefs of Staff" (Washington : GPO, 1951), 

Burma. On 1 March ABDA Command 
was formally dissolved. 

Although this first attempt at unified 
Allied command was short-lived and un- 
successful, it set a pattern which governed 
operational control of the war through its 
remaining years. This pattern amounted 
to the selection as over-all commander of 
a theater of an officer from the nation hav- 
ing the most forces in that particular 
theater. His principal subordinates were 
appointed from other nations also having 
interests and forces there. Realistically, 
the CCS tried to equate theater responsi- 
bility with national interest. On 3 March 
the Combined Chiefs approved for the 
western Pacific a new dividing line which 
cut through the defunct ABDA area. 
Burma and all Southeast Asia west of a 
north-south line between Java and Su- 
matra were added to Wavell's Indian 
command and the British Chiefs of Staffs 
were charged with the strategic direction 
of this theater. The whole Pacific east 
of the new line was given over to Am.eri- 
can JCS control. 

The Joint Chiefs divided the Pacific 
into two strategic entities, one in which 
the Navy would have paramount interests, 
the Pacific Ocean Area (POA), and the 
other in which the Army would be the 
dominant service, the Southwest Pacific 
Area ( SWPA ) . (See Map 1, Map Section 
for boundary.) Naval planners had suc- 
cessfully insisted in JCS discussions that 
all positions such as New Caledonia, the 
New Hebrides, and New Zealand which 
guarded the line of communications from 
Pearl Harbor to Australia must be con- 
trolled by the Navy. In terms of the air 
age, the JCS division of the Pacific gave 
the Army operational responsibility for 



an area of large land masses lying rela- 
tively close together where land power 
supported by shore-based air could be de- 
cisive. To the Navy the JCS assigned the 
direction of the war in a vast sea area with 
widely scattered island bases where the 
carrier plane reigned supreme. 

The American commander in the Phil- 
ippines, General Douglas MacArthur, was 
the Joint Chiefs' choice to take over direc- 
tion of SWPA operations ; Admiral Nim- 
itz was selected to head POA activities. 
Formal announcement of the new set-up 
Avas not made until MacArthur had es- 
caped from Corregidor and reached safety 
in Australia. On 18 March, wdth the con- 
sent of the Australian government, Mac- 
Arthur was announced as Supreme Com 
mander of the SWPA (CinCSWPA) 
The JCS directive outlining missions for 
both Pacific areas was issued on 30 March, 
and the confirmation of Nimitz as Com- 
mander in Chief of the POA (CinCPOA) 
followed on 3 April. By CCS and JCS 
agreement, both commanders were to have 
operational control over any force, regard- 
less of service or nation, that was assigned 
to their respective theaters. 

Nimitz still retained his command of 
the Pacific Fleet in addition to his duties 
as CinCPOA. The fleet's striking arm, 
its carriers and their supporting vessels, 
stayed under Nimitz as CinCPac no mat- 
ter where they operated. In the final 
analysis, however, the major decisions on 
employment of troops, ships, and planes 
were made in Washington with the advice 
of the theater commanders. MacArthur 
was a subordinate of Marshall and re- 
ported through him to the JCS ; an iden- 
tical command relationship existed be- 
tween Nimitz and King. 


The concern felt in Washington for the 
security of the southern route to Australia 
was acute in the days and weeks immedi- 
ately following the Pearl Hapbor attack. 
Despite world-w-ide demands on the troops 
and equipment of a nation just entering 
the war. General Marshall and Admiral 
King gave special attention to the need 
for holding positions that would protect 
Australia's lifeline. Garrison forces, most 
of them provided by the Army, moved 
into the Pacific in substantial strength to 
guard what the Allies still held and to 
block further Japanese advances. Be- 
tween January and April nearly 80,000 
Army troops left the States for Pacific 

An infantry division was sent to Aus- 
tralia to take the place of Australian 
units committed to the fighting in the 
Middle East. At the other end of the 
lifeline, a new division was added to the 
Hawaiian Island garrison. Mixed forces 
of infantry, coast and antiaircraft artil- 
lery, and air corps units were established 
in early February at Canton and Christ- 
mas Islands, southwest and south of Pearl 
Harbor. At about the same time a New- 
Zealand ground garrison reinforced by 
American pursuit planes moved into the 
Fiji Islands, and a small garrison was 
sent to the French-owned Society Islands 
to guard the eastern approaches to the 
supply route. In March a task force of 

" Unless otherwise noted the material in this 
section is derived from HqDeFor Rept of 
SamoanGru AdvB Facilities, 100ct42; 2d Mar- 
Brig AnRept, 16Jul42 ; 2d MarBrig Diary, 
23Dec41-.30Jun42 ; CG 3d MarBrig Itr to CMC, 
10Sep43; 3d MarBrig Brief of Ops, 21Mar42- 
31Aug43 ; MAG-13 War Diary, lMar42-31May43 ; 
Hist of the 7th DefBn, 21Dec42 ; Strategic 



almost division strength arrived in New 
Caledonia and the Joint Chiefs sent addi- 
tional Army garrison forces to Tongatabu 
in the Tonga Islands, south of Samoa, 
and north to Efate in the New Hebrides. 
By the end of March 1942 the supply 
route to Australia ran through a corridor 
of burgeoning island strong points and 
the potential threat of major Japanese 
attacks had been substantially lessened. 
(See Map 1, Map Section and Map 3) 

Actually the initial Japanese war plan 
contemplated no advances into the South 
Pacific to cut the line of communications 
to Australia. The Allied leaders, how- 
ever, can be forgiven for not being clair- 
voyant on this point, for the enemy's 
chance to seize blocking positions along 
the lifeline was quite apparent. Samoa 
seemed to be one of the most inviting 
targets and its tiny garrison of Marines 
wholly inadequate to stand off anything 
but a minor raid. The necessity for 
building up Samoan defenses as a prelude 
for further moves to Fiji and New Cale- 
donia had been recognized by Admiral 
King in his instructions to Nimitz to hold 
the Hawaiian-Samoa line,® and reinforce- 
ments from the States to back up those 
instructions were underway from San 
Diego by 6 January. These men, mem- 
bers of the 2d Marine Brigade, were the 
forerunners of a host of Marines who 
passed through the Samoan area and 
made it the major Marine base in the Pa- 
cific in the first year of the war. 

Only two weeks' time was necessary to 
organize, assemble, and load out the 2d 
Brigade. Acting on orders from the 
Commandant, the 2d Marine Division ac- 
tivated the brigade on 24 December at 
Camp Elliott, outside of San Diego. The 

' King'n Naval Record, 354. 

principal units assigned to the new com- 
mand were the 8th Marines, the 2d Bat- 
talion, 10th Marines, and the 2d Defense 
Battalion (dispatched by rail from the 
east coast). Colonel (later Brigadier 
General) Henry L. Larsen was named 
brigade commander. A quick estimate 
was made of the special engineering 
equipment which the brigade would need 
to accomplish one of its most important 
missions — completion of the airfield at 
Tutuila. Permission was obtained to ex- 
pend up to $200,000 in the commercial 
market for the purchase of such earth- 
moving equipment as could not be sup- 
plied from quartermaster stocks. When 
the first cargo ship arrived at San Diego 
on New Year's day, the brigade went on 
a round-the-clock loading schedule. Six- 
ty-two hours later all assigned personnel 
and gear had been loaded and the 4,798 
officers and men were on their way to 

When the news of Pearl Harbor 
reached Samoa, Lieutenant Colonel Les- 
ter A. Dessez, commanding the 7th De- 
fense Battalion, ordered his troops to man 
their positions. The Samoan Marine Re- 
serve Battalion was called to active duty 
and assigned to reinforce the defenses. 
Despite a spate of rumors and false 
alarms, no sign of the Japanese was evi- 
dent until the night of 11 January, when 
a submarine shelled the naval station for 
about seven minutes from a position 
10,000-15,000 yards off the north shore 
where the coast defense guns could not 
bear. The station suffered only light 
damage from the shells, some of which fell 
harmlessly into the bay, and two men 
were wounded slightly by fragments. 
The Marines remained on alert but re- 
ceived no further visits from the enemy. 


On 19 January radar picked up signs of 
numerous ships, and observation stations 
on the island's headlands soon confirmed 
the arrival of the 2d Brigade. 

While still at sea, General Larsen had 
received orders from the Navy Depart- 
ment appointing him Military Governoi- 
of American Samoa and giving him re- 
sponsibility for the islands' defense as well 
as supervisory control over the civil gov- 
ernment. As soon as the ships docked 
antiaircraft machine guns of the 2d De- 
fense Battalion were promptly unloaded 
and set up in the hills around Pago Pago 
harbor. The 8th Marines took over beach 
defense positions occupied by the 7th De- 
fense Battalion and immediately began 
improving and expanding them. The 
artillerymen of 2/10 and the 2d Defense 
set up their guns in temporary positions 
while they went to work on permanent 
emplacements. Navy scouting amphib- 
ians of a shore-based squadron (VS-1- 
D14) attached to the brigade soon were 
aloft on a busy schedule of antisubmarine 
and reconnaissance missions. 

The airfield on Tutuila was only 10 
per cent completed when Larsen arrived, 
but he directed that construction be 
pushed around the clock, work to go on 
through the night under lights. He also 
detailed the brigade's engineer company 
to assist the civilian contractors in getting 
the field in shape. For the 2d Brigade's 
first three months in Samoa, its days were 
filled with defense construction. There 
was little time for any combat training 
not intimately connected with the prob- 
lems of Samoan defense. The work was 
arduous, exacting, and even frustrating, 
since the brigade had arrived during the 
rainy season and the frequent tropical 
rainstorms had a habit of destroying in 

minutes the results of hours of pick and 
shovel work. 

General Larsen took immediate steps 
after his arrival in American Samoa to 
ascertain the status of the defenses in 
Western (British) Samoa, 40 or so miles 
northwest of Tutuila. On 26 January the 
brigade intelligence officer, Lieutenant 
Colonel William L. Bales, flew to Apia, 
the seat of government on the island of 
Upolu, to confer with the New Zealand 
authorities and make a reconnaissance of 
Upolu and Savaii, the two principal is- 
lands. The New Zealanders were quite 
anxious to cooperate with the Marines 
since they had a defense force of only 157 
men to guard two large islands with a 
combined coastline of over 250 miles. 
Bales, whose investigation was aimed pri- 
marily at discovering the feasibility of 
developing either or both of the islands 
into a military base, reported back that 
Upolu's harbor facilities, road net, and 
several potential airfield sites made it 
readily susceptible to base development. 
He found, on the other hand, that Savaii 
had no safe major anchorages and that its 
lava-crusted surface did "not offer air- 
field sites that could be developed quickly 
by the Japanese or anyone else." ' On 
his return to Tutuila, Lieutenant Colonel 
Bales reported to General Larsen that : 

In its present unprotected state, Western 
Samoa is a hazard of first magnitude for the de- 
fense of American Samoa. The conclusion is 
unescapable that if we don't occupy it the Jap- 
anese will and there may not be a great deal of 
time left.' 

Naval authorities in Washington and 
Pearl Harbor recognized the desirability 

'LtCol W. L. Bales Itr to CG, 2d MarBrig, 
8Feb42, Kept on Recon in Western Samoa, 8. 
"Ihid., 10. 



of occupying Western Samoa and ex- 
tended their interest to include Wallis 
l^Uea) Island, a small French possession 
320 miles from Tutuila on the western 
approaches to Samoa. Negotiations were 
entered into with New Zealand regarding 
the defense of Western Samoa, and the 
Free French government in regard to the 
occupation of Wallis. In March warn- 
ing orders were sent out to Larsen's bri- 
gade and both marine divisions to be 
prepared to furnish troops for the garri- 
soning of Western Samoa and Wallis.® 
Negotiations for the use of land and other 
facilities in Western Samoa were com- 
pleted on 20 March when Larsen and a 
New Zealand representative signed an 
agreement giving the Americans responsi- 
bility for defense of all the Samoan is- 
lands. This group, together with Wallis, 
was now considered a tactical entity and 
a new Marine brigade was to be organized 
to occupy the western islands. 

As an advance force of this new garri- 
son, the 7th Defense Battalion was sent 
to Upolu on 28 March, and a small de- 
tachment was established on Savaii. In 
the States, the 1st Marine Division at New 
River, North Carolina, organized the 3d 
Marine Brigade on 21 March with Briga- 
dier General Charles D. Barrett in com- 
mand. Its principal units were the 7th 
Marines and the 1st Battalion, 11th Ma- 
rines. The 7th's 3d Battalion and Bat- 
tery C of 1/11 were detached on the 29th 
to move overland to the west coast for 
further transfer to Samoa as part of the 
garrison for Wallis. General Larsen 
meanwhile had been directed to organize 
the 8th Defense Battalion on Tutuila, as 
the major element of the Wallis garri- 

°CMC Serial 003A7842, 20Mar42, Defense of 
Western Samoa and Wallis Island. 

son. To exercise overall authority, 
Headquarters Samoan Area Defense 
Force was established on Tutuila. Major 
General Charles F. B. Price, who was ap- 
pointed to this command, arrived with his 
staff at Pago Pago on 28 April from the 
States. On 8 May the 3d Marine Brigade 
convoy arrived off Apia and General Bar- 
rett assumed military command of West- 
ern Samoa. At the end of the month, the 
8th Defense Battalion (Reinforced) un- 
der Colonel Raphael Griffin moved into 

More than 10,000 Marine ground troops 
were stationed in the Samoan area by the 
beginning of June, and reinforcements ar- 
rived in a steady flow. Marine air was 
also well established. General Larsen's 
interest and pressure assured that Tu- 
tuila's airfield was ready for use on 17 
March, two days before the advance eche- 
lon of MAG-13 arrived. The new air 
group, organized on 1 March at San Di- 
ego, was earmarked for Price's command. 
Initially the group commander. Lieuten- 
ant Colonel Thomas J. Walker, Jr., had 
only one tactical squadron, VMF-111, op- 
erating from Tutuila's airfield, but VMO- 
151, a scout-bomber squadron, joined in 
May with the arrival of the 3d Marine 
Brigade convoy. The amphibians of the 
Navy's VS-1-D14 squadron were also put 
under Walker's command and sent for- 
ward to operate from Upolu and Wallis 
while the airfields projected for those is- 
lands were rushed to completion by naval 
construction battalions. 

Like the rest of the garrison forces in 
the South Pacific which were rushed out 
to plug a gaping hole in Allied defenses. 
General Price's defense force was never 
called upon to conduct the island defense 
for which it was organized. Samoa might 





well have become a target for enemy at- 
tacks, but the decisive Battle of Midway 
forced the Japanese to curb their soaring 
ambition.^*^ Samoa became a vast ad- 
vanced combat training camp instead of 
a battleground. Most of the units com- 
ing there after the arrival of the 2d Bri- 
gade drew heavily on the recruit depots 
for their personnel" and for these Ma- 
rines Samoan duty was an opportunity 

'" Campaigns of the Pacifio War, 3. See Part 
V, "Decision at Midway" and especially Chap- 
ter 1, "Setting the Stage — Early Naval Opera- 
tions" for events leading up to the Midway 

" At least 40% of the 3d MarBrig initial com- 
plement was straight out of boot camp. 3d Mar 
Brig AaRept, 6Sept42, 9. 

for learning the fundamentals of team- 
work in combat operations. As the need 
for defense construction was met and the 
danger of Japanese attacks lessened, Sa- 
moa became a staging area through which 
replacements and reinforcements were 
funnelled to the amphibious offensives in 
the Solomons.^^ Units and individuals 
paused for a while here and then moved 
on, more jungle-wise and combat ready, 
to meet the Japanese. 

"From December 1942 to July 1943 Samoa 
was the training center for all Marine replace- 
ment battalions raised on the east coast of the 
U. S. K. W. Condit, G. Diamond, and E. T. 
Turnbladh, Marine Corps Ch-onnd Training in 
World War II (Washington: HistBr, G-3, 
HQMC, 1956), 181-186. 


The Defense of Wake 


Wake in the Shadow of War ' 

In tlie strategic context of 1940 and 
1941, the importance of Wake, both to 
the ITnited States and Japan, was con- 
siderable. At this time the United States 
had not won its ocean-girdling net of Pa- 
cific bases, and, with the exceptions of 
Wake, Midway, and Guam, the islands be- 
tween the Hawaiians and the Philippines 
were terra incognita. Wake, a prying out- 
post north of the Marshalls and on the 
flank of the Marianas, would be a stra- 
tegic prize for Japan's ocean interests and 
a corresponding embarrassment while it 
was in the hands of the United States. 

These factors had been noted by the 
U. S. in the Hepburn Eeport of 1938 which 
recommended a $7,500,000 three-year pro- 
gram to develop the atoll as an advanced 
air base and an intermediate station on 
the air route to the Far East. Acting on 
these recommendations, initial develop- 
ment of Wake began early in 1941.- Base 
construction was given first priority, and 
by the time the first military contingent 
arrived on the atoll a civilian contractor's 
crew of approximately 1,200 men, under 
supervision of Mr. Daniel Teters, was hard 
at work. 

' For a r^sum^ of the previous history of 
Wake, see Defense of Wake, Appendix II, "Pre- 
war History of Wake, 1586-1941." Col Heinl's 
monograph has been the principal source used 
in compiling this account ; his version of the ac- 
tion has been followed closely. 

' Capt R. A. Dierdorff, USN, "Pioneer Party- 
Wake Island," U8NI Proceedings; April 1943, 

By 18 April 1941, Admiral Husband E. 
Kimmel, Commander in Chief, U. S. Pa- 
cific Fleet, became fearful that the de- 
fensive efforts had started too late. In a 
study sent to the Chief of Naval Opera- 
tions, Kimmel stressed the importance of 
Wake and asked that work on defense be 
given a higher priority than base con- 
struction. He also requested that a Ma- 
rine defense battalion be assigned to the 

In 1941 the strength of a typical defense 
battalion was 43 officers and 939 enlisted 
men, and its two most characteristic at- 
tributes were balanced structure and a 
high degree of strategic mobility. But 
mobility disappeared at the battalion's 
destination. Once its guns were in posi- 
tion, a defense battalion suffered from in- 
sufficient transportation and a shortage of 

The Pacific strategy of 1941 contem- 
plated rendering our bases relatively se- 
cure against air raids, hit-and-run surface 
attacks, or even minor landings. Fleet 
Marine Force defense battalions, or- 
ganized for defense against just such op- 
erations, could provide antiaircraft pro- 
tection, could stand off light men-of-war 
and transports, and in extreme emergency 
could fight on the beaches with individual 
weapons in the tradition that every Ma- 

' CinCPac Itr to CNO, 18Apr41. 

*USMC T/O's, D-133 through D-155-D in- 
clusive, 27Feb41; MGC Itr, 28Aug41, "Employ- 
ment of Defense Battalions." 




rine, first and last, is an infantryman.' 
Within and about the structure of such 
lightly held but secure bases, the Pacific 
Fleet would ply, awaiting the moment 
when battle could be joined with enemy 
naval forces — "to get at naval forces with 
naval forces," * as Admiral Kimmel put 
it — in decisive action for control of the 

As might be expected, the Japanese con- 
cept of strategy in the Central Pacific was 
to seize or neutralize the few advanced 
United States bases west of the Hawaiian 
Islands as quickly as possible after the 
outset of war. For this purpose Japanese 
forces in the Marshalls and Carolines (the 
Fourth Fleet) were organized along lines 
resembling an American amphibious 
force.^ Commanded by Vice Admira} 
Nariyoshi Inouye, the Fourth Fleet was 
composed of amphibious shipping, a few 
old cruisers, destroyers, submarines, shore- 
based aircraft, and a Japanese version of 
our own Fleet Marine Force: the special 
naval landing force.® Fleet headquarters 
were at Truk, where Admiral Inouye's flag 
flew in the light cruiser Kashima.^ 

The war missions of Admiral Inouye 
and his fleet had been decided generally in 
1938 when the basic East Asia war plans 

= IMd. 

" CinCPac Itr to CNO, 18Apr41, "Defense and 
Development of Wake Island." 

^ATIS (SWPA) Doe No. 17895A, "Full trans- 
lations of answers to questions concerning attack 
on Wake Island," hereinafter cited as Wake 

' The special naval landing force ( SNLF, some- 
times contracted to SLF) were Japanese Navy 
personnel organized for service and duties in 
limited land operations similar to those per- 
formed by U. S. Marines. Throughout the war, 
they gave an outstanding account of themselves. 

° Wake Attack. 

had been prepared in Tokyo.^° But it was 
not until November 1941 that detailed in- 
structions for commanders within the 
Combined Fleet were formulated and is- 
sued. In these instructions. Wake was 
dismissed in a single phraSe : 

Forces of the Fourth Fleet : 
Defend the South Seas Islands, patrol, main- 
tain surface communications, capture 
Wake . . . ." 

Wake would be strictly a local opera- 
tion. By Admiral Inouye's scheme, 450 
special naval landing force troops could, 
in a pinch, turn the trick." 


On 23 June 1941 the Chief of Naval 
Operations directed that elements of the 
1st Defense Battalion, FMF, be estab- 
lished at Wake "as soon as practicable." 
This directive (as eventually modified) 

"'USSBS(Pac), NavAnalysisDiv, Interroga- 
tions of Japanese Officials, 2 vols (Washington: 
GPO, 1946), "Japanese Naval Planning," I, 176, 
hereinafter cited as VSSB8 Interrogations with 
subject or interviewee. 

" Campaigns of the Pacific War, 47. 

" U8SBS Interrogations, "Japanese Capture of 
Wake Island," II, 371, hereinafter cited as Cap- 
ture of Wake. 

"Unless otherwise noted the material in this 
section is derived from CO 1st DefBnDet Wake, 
Rept to CMC, 18Mar46, hereinafter cited as 
Devereujc Rept; Col P. A. Putnam Rept to CMC, 
180ct45, hereinafter cited as Putnam Rept; in- 
formal reports by key subordinates to Cols Dev- 
ereux and Putnam on which the official reports 
are largely based, hereinafter cited as (officer's 
name) Rept; ships' logs of the U. S. naval ves- 
sels concerned ; Col J. P. S. Devereux, The Story 
of Wake Island { Philadelphia : J. P. Lippincott 
Company, 1947), hereinafter cited as Devereux 





specified that the following units should 
compose the defensive garrison : 

Four 3-inch antiaircraft batteries 
Tliree 5-incii seacoast batteries 
Appropriate automatic weapons 
One SCR-268 fire-control radar, and one SCR- 
270B search radar.'* 

CNO's "as soon as practicable" was 
translated into immediate action by the 
Pacific Fleet. About 1 August Major 
Lewis A. Hohn with five officers and 173 
enlisted Marines and sailors from the 1st 
Defense Battalion commenced loading the 
USS Regulus, a twenty-year-old "Hog Is- 
land'' transport which would carry the 
battalion advance detail to Wake. Regn- 
lus sailed on 8 August, and arrived ofi^ 
Wake on 19 August. Weapons and camp 
equipment were lightered ashore, and by 
the time the Regulus departed on 22 Au- 
gust, a camp facing the lagoon had been 
set up on a site near the west end of 
Wake's west leg. To distinguish this 
camp from the one west of Heel Point 
housing the 1,200 Pacific Naval Air Base 
contract workmen, the Marine camp was 
designated as Camp One. The civilian 
establishment became known as Camp 
Two. (See Map 4) 

Wake, as it appeared to the Marines of 
the 1st Defense Battalion, was a V-shaped 
atoll composed of three islands : Wake Is- 
land proper," the body of the V; and 
Wilkes and Peale, the two tip-ends. Its 
land mass consisted of some 2,600 acres 
of sand and coral. Offshore, heavy surf 
roared continually against a coral reef 
which surrounded the whole atoll at dis- 

tances varying from 30 to 1,000 yards. 
The beaches and much of the terrain in- 
land were covered with coral boulders, 
some large enough to conceal several men. 
The interior lagoon, although affording 
sufficient surface and depth for seaplanes, 
was studded with coral heads and foul 
ground which had to be dredged before 
ships could enter the single channel be- 
tween Wilkes and Wake Island. Despite 
Wake's limited land area, its coastline ex- 
ceeded 21 miles. An excellent vignette 
of Wake in 1941 was given by Colonel 
Bayler : 

Wake is by no means the bare sandy spit one 
thinlis of when atolls are mentioned. Consid- 
erable areas of it are covered by woods, and 
though the trees are small, their thick foliage 
and the scrubby tangled underbrush provided 
admirable cover . . . Walking in these jungles 
was difficult but not impossible . . .*'' 

In August 1941, Wake was in rapid 
transition from its past solitude to the 
mechanized modernity of an outlying air 
base. Patrol plane facilities and a con- 
crete ramp, the result of Pan American's 
pioneering, were already available on 
Peale." Just inshore of Peacock Point 
along the west leg of Wake Island a nar- 
row airstrip, 5,000 by 200 feet, had been 
chopped out of the dense growth. A 
main roadnet of packed coral was taking 
shape rapidly as the contractor's work- 

"CNO Itr to CinCPac, 23Jun41, "Establish- 
ment of defensive garrison on Wake Island." 

'^ To prevent confusion, Wake Island, as dis- 
tinguished from the entire atoll, will herein- 
after be entitled "Wake Island," whereas the 
single word, "Wake" will designate the atoll. 

'"LtCol W. L. J. Bayler, hast Man off Wake 
Island (Indianapolis, Bobbs-Merrill, 1913), 62, 
hereinafter cited as Last Man off Wake Island. 

" In 1935, with Navy cooperation. Pan Ameri- 
can Airways began development of a staging and 
refueling base on Peale to service its big clippers 
on the run to the Orient. At the time of this 
narrative major facilities included, in addition 
to those mentioned above, a powerful radio sta- 
tion, a pier, and a small but excellent hotel for 
overnight accommodation of passengers. Dier 
dorflf, op. cit., 501. 



men blasted, slashed, and dozed the ter- 
rain of Wake. 

In spite of the need for haste, rigid 
official separation existed between the con- 
struction efforts of Marines and those of 
the contractors." Operating on a semi- 
private basis with their heavy equipment, 
supplies, and facilities the naval air base 
contract workers were concerned with 
building roads, shops, utilities, quarters, 
air base facilities, and the like. They 
built no defense installations. This con- 
struction fell solely to the Marines who 
had little engineering equipment except 
picks and shovels or the infrequent lux- 
ury of a borrowed civilian bulldozer. 
The Marines installed their heavy weap- 
ons by hand, hewed emplacements and 
foxholes from the coral, and erected their 
own living quarters. Understanding this 
basic difference in available means, the 
Navy's construction representative, Lieu- 
tenant Commander Elmer B. Greey," and 
the civilian general superintendent, Mr. 
Daniel Teters, did their best to assist the 
shorthanded and meagerly equipped Ma- 
rines. At no time, even after the out- 
break of war, did the contractor's estab- 
ishment or workmen come under full mil- 
itary control. 

On 15 October Major Hohn was relieved 
as Marine detachment commander by Ma- 
jor James P. S. Devereux, who until this 
time had been executive officer of the 1st 
Defense Battalion. Major Devereux also 

" Capt W. S. Cunningham, USN, transcript of 
recorded interview, "History of Wake Island De- 
fense," 9Jan46, 3, hereinafter cited as Cunning- 
ham Interview. 

" "Resident Offlcer-in-Charge" was LCdr 
Greey's oflBcial designation. With four enlisted 
Navy radiomen to maintain his communications, 
he was, until the arrival of Maj Hohn's de- 
tachment, sole naval representative on Wake. 

became Island Commander, an additional 
duty which he would hold until relieved 
late in 1941 by a naval officer. Commander 
W. S. Cunningham, at this time still navi- 
gator of the USS Wright. 

Major Devereux, as he saw Wake at this 
time, describes it as follows : 

When I arrived on the island, the contractor's 
men working on the airfield near the toe of Wake 
proper had one airstrip in usable condition and 
were beginning the cross-runway. Five large 
magazines and three smaller detonator maga- 
zines, built of concrete and partly underground, 
were almost completed in the airfield area. A 
Marine barracks, quarters for the Navy fliers 
who would be stationed on the island, ware- 
houses and shops also were going up on Wake. 
On Peale Island, work was progressing on a 
naval hospital, the seaplane ramp and parking 
areas. On Wilkes, there were only fuel storage 
tanks and the sites of proposed powder maga- 
zines, but a new deepwater channel was being 
cut through the island. In the lagoon, a dredge 
was removing coral heads from the runways for 
the seaplanes which were to be based at Wake. 
Some of these installations were nearly finished ; 
some were partly completed ; some were only in 
the blueprint stage.* 

To bring Wake's defenses to the high- 
est possible state of readiness in the short- 
est time. Major Devereux found much to 
be done. In addition, as senior repre- 
sentative of the armed forces on Wake, he 
was confronted by other demanding prob- 
lems. To reinforce Army air strength in 
the Philippines, B-17 "Flying Fortresses" 
w^ere being staged across the Pacific ^^ 
through Wake, but no aviation ground 
crews were available there to service the 
big airplanes. Some 3,000 gallons of gas- 
oline for each of these planes therefore 
had to be manhandled and hand-pumped 

'"Devereux Story, 25. 
" War Reports, 67. 



by the Marines.^^ This they did in addi- 
tion to their normal duties, and the fueling 
tasks came at all hours of the day or night. 
It was ironic that many of these aircraft, 
which cost Wake so many man-hours of 
vital defensive preparations, would be 
trapped on the ground by the initial Japa- 
nese attacks on Clark and Nichols Fields 
in the Philippines. 

Although this servicing of Army planes 
represented the heaviest single additional 
duty imposed upon the Marines, they were 
also required to act as stevedores in the 
time-consuming and exhausting process of 
unloading ships which arrived at the atoll. 
This work was required until the channel, 
berthing and turning facilities inside the 
lagoon could be completed. These addi- 
tional duties hampered defense work dur- 
ing the autumn of 1941; but fortunately 
the detachment needed little combat train- 
ing because it contained a number of "old 
Marines"' of the best type.^^ On 2 No- 
vember, two weeks after Major Devereux's 
arrival, the Wake garrison was augmented 
by a draft from the parent 1st Defense 
Battalion. This group included 9 officers 
and 200 enlisted men who arrived from 
Pearl on board the USS Castor. This 
brought the total Marine strength on 
Wake to 15 officers and 373 enlisted 

^^ Tankers would pump bulk aviation gas into 
tank storage ashore ; Marine working parties 
would pump this gasoline into 50-gallon drums 
and transfer the drums to dispersed fuel dumps ; 
finally, on arrival of planes the same gasoline 
would again be pumped by the same means into 
a lone tank-truck for delivery to the aircraft. 
When time pressed — as it usually did — Marines 
reinforced the truck by pumping directly from 
50-gallon drums into the Fortresses. 

^ Devereuu) Story, 27. 

During October and November prog- 
ress on and about the airstrip, by now 
a going concern, indicated that there was 
room on Wake for the aviation compo- 
nent of fighters necessary to balance and 
round out the defense force. Commander, 
Aircraft Battle Force, had determined 
that this was to be Marine Fighter 
Squadron 211, supported in its independ- 
ent role by a provisional service detach- 
ment drawn from Marine Air Group 21, 
to which VMF-211 was assigned. To es- 
tablish the ground facilities required to 
maintain this squadron. Major Walter 
L. J. Bayler from the staff of MAG-21, 
together with a detachment of 49 Ma- 
rines commanded by Second Lieutenant 
Robert J. Conderman,^* were dispatched 
from Pearl on 19 November in USS 
Wright, an aircraft tender which was also 
bringing out the prospective Island Com- 
mander and commanding officer of the 
Naval Air Station. 

While the Wright plowed westward 
bearing VMF-211's ground components, 
the air echelon of that squadron, consist- 
ing of the squadron commander, nine of- 
ficers and two enlisted pilots,^^ had on the 

"This detachment, like a similar one organ- 
ized for the Marine air component at Midway, 
had been provisionally made up from key per- 
sonnel representing each squadron in MAG-21, 
inasmuch as, at the time of organization, firm 
decision had not been made as to which squad- 
rons from that group would be assigned to which 
islands. Wake aviation's ground detachment, 
therefore, included personnel not only from 
VMF-211 but from H&S Sq-21 and VMSB-231 
and -232. CO MAG-21 Rept to CMC, 23Dec41. 

'^The pilots of VMF-211's Wake detachment 
were: Maj Paul A. Putnam (commanding), 
Capts Henry T. Elrod, Herbert C. Frueler, Frank 
C. Tharin ; IstLt George A. Graves ; 2dLts Rob- 
ert J. Conderman (in command of advance 
detail and ground maintenance, but also a pi- 
lot), Carl R. Davidson, Frank J. Holden, John 



afternoon of 27 November received secret 
verbal warning orders to prepare for em- 
barkation on board a carrier. Such or- 
ders had been expected by the squadron 
commander (though not by the pilots, vir- 
tually all of whom carried little more 
than toilet articles and a change of cloth- 
ing) , and few preparations were required. 
The squadron had only to fly the 12 new 
F4F-3 (Grumman Wildcat) fighters 
from Ewa Mooring Mast (as that air sta- 
tion was then designated) over to Ford 
Island, the naval air base in the middle 
of Pearl Harbor, for further transfer by 
air to the flight deck of the USS Enter- 
prise. This was a routine operation for 
Marine pilots, and except for their un- 
familiarity with the new aircraft, and the 
fact that one plane's starter misbehaved,^" 
the morning flight of 28 November onto 
the Enter jyrise went off without incident." 

The best description of VMF-211's voy- 
age to Wake is contained in a personal 
letter, composed on the eve of the squad- 
ron's debarkation, from Major Paul A. 
Putnam to Colonel Claude A. Larkin who 
commanded MAG-21. Excerpts are 
quoted : 

At Sea, 
December 3, 1941. 
Dear Colonel Larkin : 

It is expected that we will go ashore to- 
morrow morning. The extreme secrecy under 
which we sailed is still in effect and I under- 

F. Kinney, David D. Kliewer, Henry G. Webb ; 
TSgt William J. Hamilton, and SSgt Robert O. 

''"A hint as to the importance of the squad- 
ron's mission might have been drawn at this 
time from the fact that, when this starter trou- 
ble developed, the pilot of this defective plane 
was flown by a torpedo plane to the carrier 
where a brand-new F4F-3 from an Enterprise 
squadron was issued to him. 

"Maj. P. A. Putnam Itr to CO ]VIAG-21, 

stand is to remain so at least until this Force 
has returned to Hawaiian operating area. 
Therefore I am sending this first report via guard 
mail on this ship, rather than by air mail after 
landing . . . 

You will recall that I left one plane at Ford 
Island. The Admiral at once gave me a plane 
to replace it, from VF-6 ; and he made it plain 
to me and to the whole ship that nothing should 
be overlooked nor any trouble spared in order 
to insure that I will get ashore with 12 air- 
planes in as near perfect condition as ptossible. 
Immediately I was given a full complement of 
mechs and all hands aboard have continually 
vied with each other to see Wiho could do the 
most for me. I feel a bit like the fatted calf 
being groomed for whatever it is that happens 
to fatted calves, but it surely is nice while it 
lasts and the airplanes are pretty sleek and 
fat too. They have of course been checked and 
double checked from end to end, and they have 
also been painted so that all 12 are now of 
standard blue and gray . . . 

The Admiral seems to be most determined to 
maintain secrecy regarding the position and ac- 
tivity of this Force. There has been a continu- 
ous inner air patrol during daylight, and a full 
squadron has made a long search to the front 
and flanks each morning and evening. They are 
armed to the teeth and the orders are to attack 
any Japanese vessel or aircraft on sight in order 
to prevent the discovery of this Force. 

My orders, however, are not so direct. In fact 
I have no orders. I have been told informally 
by lesser members of Staff that I will be given 
orders only to fly off the ship and go to the land, 
and that there will be nothing in the way of 
instructions other than to do what seems ap- 
propriate at the moment. Of course I shall go 
and ask for orders and instructions, but it 
seems unlikely that I shall be given anything 
definite . . 

This is written Wednesday forenoon. Should 
I receive any orders at variance with the fore- 
going, I will add a postscript. Otherwise I 
think of nothing further of importance or in- 
terest at this time. . . . 

When the Enterprise had reached a 
point approximately 200 miles northeast 
of Wake, the squadron, from a materiel 
standpoint, was "as far as possible ready 



for combat service," according to Major 
Putnam. However, he added, it was: 

. . . seriously handicapped by lack of experi- 
ence in the type of airplane then used. It is 
believed that the squadron was excellently 
trained and well qualified for war duty in a 
general sense, but it was unfortunate that the 
new type of airplane, so radically different from 
the type in which training had been conducted, 
had been received too recently to permit famil- 
iarization in tactical flying and gunnery.^* 

On the morning of 4 December this force 
was met by a Navy PBY sent out from 
Wake,2« and the VMF-211 aircraft took 
off from the Enterprise and followed this 
plane to the atoll. Within less than two 
hours the last F4F-3 had pancaked on the 
narrow strip at Peacock Point. 

Major Bayler had arrived on 29 No- 
vember and already was busy setting up 
airbase communication facilities. Com- 
mander Cunningham had succeeded Ma- 
jor Devereux as Island Commander, and 
Lieutenant Conderman and his 49 head- 
quarters and service personnel were wait- 
ing to greet the squadron, but the aircraft 
operating facilities at Wake were hardly 
in a finished stage. The landing strip, 
although sufficient in length, was too nar- 
row to permit safe operation of more than 
one airplane at a time. Takeoffs or land- 
ings by section were thus impossible. 
Parking was extremely restricted, and all 
areas about the hardstand mat were in 
such rough and unfinished condition that 

^ Putnam Rept, 13. 

"' On the day before, to the surprise of the men 
on Wake, a 12-plane squadron of PBY's had 
glided down onto the lagoon, anchored, and 
commenced a daily series of long-range air 
searches to the south of Wake. These seaplanes, 
however, were recalled from Wake on 5 Decem- 
ber. The PBY which assisted VMF-211 with its 
navigation was from this squadron. Last man 
off Wake Island, 29. 

passage of airplanes over them, even when 
pushed by hand, could cause serious plane 
damage. Fueling still depended on hand 
pumps and man power. No shelters or 
aircraft revetments existed, and the new 
planes were somewhat puzzling to pilots 
and mechanics who had no instruction 
manuals. Major Putnam began imme- 
diately to negotiate for the construction 
of revetments,^" and he also began a train- 
ing program to be carried on in conjunc- 
tion with the daily dawn and dusk pa- 
trols which started on the morning after 
VMF-211 arrived. 

These patrols, executed by four air- 
craft, circled the atoll approximately 50 
miles out, and pilots combined this duty 
with navigation and instrument training. 
Instrument practice was particularly im- 
portant because Wake had no electronic 
homing or navigational aids suitable for 
fighter operations, and the atoll was a 
small mark for pilots to locate through a 
floor of intermittent clouds.^^ 

Other changes had taken place since the 
arrival of the Wright. Commander 

*" "Backed by a written request from the Com- 
mander, Aircraft Battle Force, a request was 
made through the Island Commander to the Ci- 
vilian Contractor's superintendent on the morn- 
ing of 5 December, asking for the immediate 
construction of bunkers for the protection of 
aircraft, and outlining various other works to 
follow. Great emphasis was put on the fact 
that speed, rather than neatly finished work, 
was required. However, an inspection that 
afternoon revealed a young civil engineer la- 
boriously setting out stakes with a transit and 
three rodmen. It required an hour of frantic 
rushing about and some very strong language 
to replace the young engineer and his rodmen 
with a couple of Swedes and bulldozers." Put- 
nam Rpt, 6. 

" HistSec, HQMC interview with IstLt J. F. 
Kinney, 23Jul45, 4, hereinafter cited as Kinney 



Cunningham had brought with him Com- 
mander Campbell Keene, eight Navy of- 
ficers, and 58 bluejackets who comprised 
the initial detachment of the Naval Air 
Station. All these personnel, like the 
Army Air Force communication detach- 
ment ^^ of one officer and four soldiers, 
were without arms or field equipment. 
In spite of the efforts, men, and equipment 
consigned to Wake, the situation was 
still grim on 6 December 1941. The 
ground defenses, embodying the complete 
artillery of a defense battalion, had been 
emplaced during 12-hour working days, 
and some protective sandbagging and 
camouflage accomplished. But to man 
these weapons the 1st Defense Battalion 
detachment had only 15 officers and 373 
enlisted men, although the 1941 T/0 
called for 43 officers and 939 men. This 
meant that one 3-inch antiaircraft bat- 
tery ^^ w^as entirely without personnel, and 
that each of the other two batteries could 
man only three of its four guns. Thus 
only six of the twelve 3-inch guns on the 
island could be utilized. Only Battery 
D had its full allowance of fire-control 
equipment. Battery E had a director but 
no height finder, and it had to get alti- 
tude data by telephone from Battery D. 
There were not half enough men to em- 
ploy the ground and antiaircraft machine 
guns. There was no radar, despite plans 
for its eventual provision, and the search- 
light battery did not have sound locators 
with which to detect approaching air- 

'^ Commanded by Capt Henry S. Wilson, USA. 
This detachment manned an Army Airways 
Communication Service radio van to assist fi- 
ll's en route westward. 

^This was Btry F. For this battery, how- 
ever, the necessary fire control equipment had 
not yet arrived ; so, even with full gun crews, 
its effectiveness would have been slight. 

craft. Only the crews of the 5 -inch sea- 
coast batteries were at or near authorized 
strengths, and they also were devilled by 
unending minor shortages of tools, spare 
parts, and miscellaneous ordnance items.^* 

Peale Island's base development and de- 
fensive organization were the most ad- 
vanced in the atoll. Although Battery 
B, the 5-inch seacoast unit at Toki Point, 
had been fully organized only after the 
arrival of personnel on 2 November, its 
position was in good shape. Much the 
same could be said of Battery D, 3-inch 
antiaircraft, set up near the southeast end 
of the island. All emplacements had not 
been completely sandbagged, but there 
were adequate personnel shelters plus un- 
derground stowage for 1,400 rounds of 
3-inch ammunition. Telephone lines, al- 
though not buried, linked all positions 
with the island command post. Work on 
Wake Island was not far behind. Bat- 
tery A, the 5-inch seacoast unit at Pea- 
cock Point, was completely emplaced and 
well camouflaged although it lacked indi- 
vidual shelters. Battery E, (3-inch anti- 
aircraft) , although working with only 43 
Marines, had completely emplaced, sand- 
bagged and camouflaged two guns and the 
director, and work on the third gun was 
nearly completed by 6 December. Tele- 
phone lines (with important trunks 
doubled or tripled) connected all units on 
Wake Island, but the wire was on the 

"Wilkes Island was the least devel- 
oped," reported Captain Wesley McC. 
Piatt, the local commander: 

... At the outbreak of war, weapons . . . 
had been set up. All were without camouflage 
or protection except the .50 caliber machine 

** File, dispatches received from Wake, 7-23 
Dec41, hereinafter cited as Wake File. 



guns, which had been emplaced. All brush east 
of the new channel had been cleared. The re- 
maining brusli west of the new channel was 
thick and ... as a result of . . . this [the] 
... .50 caliber machine guns had been placed 
fairly close to the water line. The beach it- 
self dropped abruptly from 21/2 to 4 feet just 
above the high water mark.*^ 

In addition to four .50 caliber AA and 
four .30 caliber machine guns, Piatt had 
two searchlights and one 5-inch seacoast 
battery (L) which was set up at Kuku 
Point. The four 3-inch guns destined for 
Battery F were parked on Wilkes with- 
out personnel or fire control gear. Wire 
communications were in between the is- 
land command post and all units.^" 

Wake, intended primarily as a patrol 
plane base for PBY's, "the eyes of the 
Fleet," had no scouting aircraft after the 
PBY's departed on 5 December, and only 
the most primitive facilities for any type 
of aircraft operations. Its defending 
fighter squadron was learning while 
working, and these planes had neither 
armor nor self-sealing fuel tanks. In ad- 
dition, their naval type bomb racks did 
not match the local supply of bombs.^^ 

Exclusive of the 1,200 civilian contract 
employees, the military population of 
Wake (almost twenty per cent of whom 
were without arms or equipment) totalled 
38 officers and 485 enlisted men : ^^ 

=^LtCol W. McC. Piatt reply to HistSec, 
HQMC questionnaire, 10Mar47. 

"^ Ibid., 2. 

'' Capt Frueler, squadron ordnance officer, at 
this moment was devising homemade modifica- 
tions of the troublesome bomb lugs. By 8 De- 
cember two 100-pound bombs could be precari- 
ously swung onto each aircraft, though hardly 
in any manner to inspire pilot confidence in 
clean release or assurance that return to base 
could be accomplished without dangling armed 

"' Devereux Rept. 

1st Defense Battalion 1.5 officers, 373 enlisted 

detachment : 
VMF-211 and attach- 12 officers, 49 enlisted 

ments : 
U. S. Naval Air Sta- 10 officers, 58 enlisted 

tion : (without arms). 

Army Air Corps : 1 officer, 4 enlisted 

(without arms). 
USS Triton: 1 enlisted (without 

arms, landed for 
medical attention). 

Thus there were only 449 Marines on the 
atoll who were equipped and trained for 

Supplies on Wake, although aggravat- 
ingly short in many particular items, 
were generally adequate. The Marines 
had a 90-day supply of rations, and the 
civilian workers had a six-month supply. 
No natural water supply existed, but a 
sufficient number of evaporators were in 
service. Ammunition and aviation ord- 
nance supplies initially could support 
limited operations, but would not with- 
stand a protracted defense. Medical sup- 
plies were those normal for a remote, 
outlying station and could thus be con- 
sidered adequate.^® In addition to the 
naval medical equipment and personnel on 
Wake, the contractor's organization oper- 
ated a fully-equipped hospital in Camp 

But since November, when dispatches 
had warned that the international situa- 
tion demanded alertness, the atoll was as 
ready for defense as time and material 
available permitted. When this warning 
arrived. Major Devereux, then the island 
commander, asked whether the civilian 
workers should be turned to tasks dealing 
more directly with military defense, but 
he was told not to revise work priorities. 
Small-arms ainmunition was nevertheless 

' Maj. W. L. J. Bayler Rept, 9-10. 
° Cunningham Interview, 3. 



issued to individual Marines, and ready - 
service ammunition was stowed at every 
gun position. A common "J"-line (so- 
called) which augmented normal tele- 
phone circuits, joined all batteries, com- 
mand posts, observation posts, and othei' 
installations with which the commander 
might need contact during battle,^^ and 
primitive "walky-talkies" formed a radio 
net established to parallel wire communi- 
cations between command posts on Wake 
Island, Wilkes, and Peale. Atop the 50- 
foot steel water tank at Camp One, the 
highest point on Wake, Major Devereux 
had established a visual observation post 
linked by field telephone to the command 
post. This OP, with a seaward horizon of 
about nine miles, was the only substitute 
for radar. 

On the morning of Saturday, 6 Decem- 
ber, Major Devereux found time to hold 
the first general quarters drill for the en- 
tire defense battalion. "Call to Arms" was 
sounded, and all gun positions were man- 
ned (to the extent which personnel short- 
ages permitted), communications tested, 
and simulated targets were "engaged." *^ 
The drill ran smoothly, and Major Deve- 
reux granted his men an almost unheard- 
of reward : Saturday afternoon off, and 
holiday routine for Sunday. 

His timing of this "breather" was better 
than he knew. 

Maj W. L. J. Bayler Kept, 3. 

" Prior to the outbreak of war, no opportunity 
liad been found for test firings, calibration, or 
other gunnery exercises after emplacement of 
weapons on Wake. The first actual firing was 
in combat against the Japanese. Cunningham 
Interview 3. 

The Enemy Strikes ' 


The Pan American Airways Philippine 
Clipper which had spent the night of 7-8 
December at Wake re-embarked passen- 
gers shortly after sunrise on Monday ^ 
8 December, taxied into the calm lagoon, 
and soared toward Guam. Ashore break- 
fast was nearly over, and some Marines 
were squaring away their tents prior to 
falling out for the day's work. Major 
Devereux was shaving. In the Army Air- 
ways Communications Service radio van 
near the airstrip, an operator was coming 
up on frequency with Hickam Field on 
Oahu when at 0650 a frantic uncoded 
transmission cut through : Oahu was under 
enemy air attack. 

Captain Henry S. Wilson snatched the 
message and rushed to Devereux's tent. 
The major tried unsuccessfully to reach 
Commander Cunningham by telephone, 
and then called the base communication 
shack. There, a coded priority ^ transmis- 
sion from Pearl was being broken down. 
Devereux put down the telephone and 
ordered the field music to sound "Call to 

^ Unless otherwise noted, the material in chap 
2 is derived from Devereux Rept; Putnam Kept; 
{officer's name) Repts; Buyler Rept; Devereux 

" By east longitude date ; this was the same as 
Sunday, 7 December east of the date line. 

' At this time relative priorities in dispatch 
traffic were as follows: Urgent (to be used only 
for initial enemy contact reports). Priority, 
Routine, Deferred. Thus a priority dispatch 
presented a considerably more important trans- 
mission than it now would. 

Arms.* Gunnery sergeants broke out their 
men and made sure that all had their am- 
munition. The Marines then piled into 
trucks which rushed them to the battery 
areas. By 0735 all positions were manned 
and ready, the planned watch was estab- 
lished atop the water tank in Camp One, 
and defense battalion officers had held a 
brief conference. 

The dawn air patrol was up before the 
news came from Pearl,^ but aviation per- 
sonnel took hurried steps to safeguard the 
new Wildcats still on the ground. The 
Philippine Clipper was recalled ten 
minutes after its takeoff, and it circled 
back down to the lagoon. But in spite of 
these measures, things were not running 
smoothly at the airstrip. VMF-211 had 
been on Wake only four days and could 
hardly call itself well established. Air- 
craft revetments still being dozed would 
not be ready until 1400 that day, and suit- 
able access roads to these revetments like- 
wise were unfinished. Existing parking 
areas restricted plane dispersal to hazard- 
ously narrow limits. As Major Putnam 
stated it : 

The Squadron Commander was faced with a 
choice between two major decisions, and in- 
evitably he chose the wrong one. Work was 

* Cdr Cunningham, who immediately recalled 
the Philippine Clipper, has since stated that it 
was he who ordered the defense battalion to gen- 
eral quarters, but it appears that this action had 
already been taken prior to his issuance of any 
order. Cunningham Interview, 4. 

^ Kinney Interview, 3. 




progressing simultaneously on six of the pro- 
tective bunkers for the airplanes, and while 
none was available for immediate occupancy, all 
would be ready not later than 1400. Protection 
and camouflage for facilities were not available 
but could be made ready within 24 hours. Eox- 
holes or other prepared positions for personnel 
did not exist but would be completed not later 
than 1400. To move the airplanes out of the 
regular parking area entailed grave risk of 
damage, and any damage meant the complete 
loss of an airplane because of the complete ab- 
sence of spare parts . . . The Squadron Com- 
mander decided to avoid certain damage to his 
airplanes by moving them across the rough 
ground, to delay movements of material until 
some place could be prepared to receive it, and 
to trust his personnel to take natural cover if 

Thus VMF-211's handful of pilots and 
mechanics spent the morning dispersing 
aircraft as widely as possible in the usable 
parking area, relocating the squadron 
radio installation from its temporary site 
to a covered one, and arming and servic- 
ing all aircraft for combat. 

At 0800, only a few hours after the 
blazing and dying AHzona had broken out 
her colors under enemy fire at Pearl Har- 
bor, Morning Colors sounded on Wake. 
Defensive preparations hummed. Trucks 
delivered full allowances of ammunition 
to each unit, the few spare individual 
weapons in Marine storerooms were 
spread as far as they would go to the un- 
armed Air Corps soldiers and Naval blue- 
jackets, and gas masks and helmets of 
World War I vintage were distributed to 
the battery positions. Watches were set 
at fire control instruments and guns, while 
the balance of personnel worked on fox- 
holes and filled the few remaining sand- 
bags. The 3-inch antiaircraft batteries 
were specifically directed to keep one gun, 
plus all fire control instruments, fully 

manned. Marine units and the Island 
Commander hastily set up command posts. 
Commander Cunningham located his CP 
in Camp Two, and VMF-211's remained 
in tlie squadron office tent. Aviation per- 
sonnel had to stick with their jobs of belt- 
ing extra ammunition and transferring 
bulk fuel into more dispersable drums. 

At 0900 the four-plane combat air patrol 
returned to base. Tlie planes were re- 
fueled while the four pilots ^ took a smok- 
ing break, and then clambered back into 
F4F's 9 through 12 and took off again to 
scout the most likely sectors for enemy 
approach. Shortly after this the pilot of 
the Philippine Clipper, Captain J. H. 
Hamilton, reported for duty to Major 
Putnam at VMF-211's headquarters. He 
had orders from the Island Commander 
to make a long-range southward searcl\ 
with fighter escort. These orders, how- 
ever, were later cancelled.* 

While VMF-2irs combat air patrol 
made a swing north of Wake at 12,000 
feet, 36 twin-engined Japanese bombers 
were flying northward toward the atoll. 
This was Air Attack Force No. 1 of the 
Twenty-Fourth Air Flotilla, based at Roi, 
720 miles to the south.® As the enemy 
group leader signalled for a gliding let- 

° Putnam Rept, 8. 

' There were : Capt Elrod, who had relieved 
Maj Putnam on station, and 2dLt Davidson in 
one section, and IstLt Kinney and TSgt Hamil- 
ton in the other. 

* Orders were changed and the clipper took 
off for Midway at 1250 that afternoon to evacu- 
ate certain PAA personnel plus all passengers. 
Mr. H. P. Hevenor, a government official who 
missed the plane, was marooned on Wake and 
eventually ended up in Japanese hands. "It 
struck me as a rather drastic lesson in the 
wisdom of punctuality," commented Col Deve- 
reux. Devereux Story, 58. 

° Notes on Enemy Interviews, n. d., hereinafter 
cited as Enemy Notes. 



down in his 10,000-foot approach, he noted 
that the south coast of the atoll was 
masked by a drifting rain squall at about 
2,000 feet. The three Japanese divisions, 
in 12-plane Vs, dropped rapidly down into 
the squall and emerged a few seconds later 
almost on top of the Wake airstrip. 
First Lieutenant William W. Lewis, com- 
manding Battery E at Peacock Point, saw 
these planes at 1150, and he grabbed a 
"J"-line telephone to warn Devereux. 
Just as the major answered, a spray of 
bright sparks began to sail through the air 
ahead of the enemy formation. One civil- 
ian thought "the wheels dropped off the 
airplanes." But the planes had not come 
to lose their wheels. Japanese bombs were 
falling on Wake. 

Lewis, an experienced antiaircraft 
artilleryman, liad not only complied with 
the commanding officer's directive to keep 
one gun manned, but had added another 
for good measure. Within a matter of 
seconds he had two of Battery E's 3-inch 
guns firing at the Japanese,^" and .50 
caliber guns along the south shore of 
Wake quickly took up the fire. A tight 
pattern of 100-pound fragmentation 
bombs and 20mm incendiary bullets struck 
the entire VMF-211 area where eight 
Grummans were dispersed at approxi- 
mately hundred-yard intervals. While 
two 12-plane enemy divisions continued 
to release bombs and to strafe Camp 
Two, one division broke off, and swung 
back over Camp One and the airstrip. 

" Battery E, it will be recalled, had no height 
finder but was supposed to rely for this data 
on telephonic information from Battery D on 
Peale. Without waiting for word from Peale, 
Lt Lewis made a quick estimate of target alti- 
tude, cranked it onto his director, and had the 
battery in action within a matter of seconds. 

For a second time within less than ten 
minutes the airstrip was bombed and 
strafed. By 1210 the strike was over. The 
enemy planes turned away and com- 
menced their climb to cruising altitude. 
"The pilots in every one of the planes were 
grinning widely. Everyone waggled his 
wings to signify '■Banzai'. ^^ " 

The enemy attack burned or blasted 
seven of the eight F4F-3's from propeller 
to rudder, and the remaining Wildcat sus- 
tained serious but not irreparable dam- 
age to its reserve fuel tank. A direct 
bomb hit destroyed Major Bayler's air- 
ground radio installation, and the whole 
aviation area flamed in the blaze from 
the 25,000-gallon avgas tank which had 
been hit in the first strike. Fifty-gallon 
fuel drums burst into flame. VMF-211's 
tentage, containing the squadron's scanty 
stock of tools and spares, had been riddled 
and partially burned. Worst of all, 23 of 
the 55 aviation personnel then on the 
ground were killed outright or wounded 
so severely that they died before the fol- 
lowing morning, eleven more were woun- 
ded but survived. At one stroke, VMF- 
211 had sustained nearly 60 per cent 
casualties. Nearly 50 per cent of the 
ground crewmen were dead. Three pilots 
(Lieutenants George A. Graves, Robert J. 
Conderman, and Frank J. Holden) were 
killed, and another, Lieutenant Henry G. 
Webb, was seriously wounded. Three 
more pilots. Major Putnam, Captain 
Frank C. Tharin, and Staff Sergeant 
Robert O. Arthur, had received minor 
wounds but remained on duty. In Camp 

"Account by Norio Tsuji, a Japanese observer 
during the raid. ATIS (SWPA), Enemy Pub- 
lications No. 6, "Hawaii-Malaya Naval Opera- 
tions," 27Mar43, 27-38, hereinafter cited as 
Hawaii-Malaya NavOps. 



Two and the adjacent Pan American area, 
the hotel and other seaplane facilities 
were afire, the Philippine Clipper had re- 
ceived a few stray machine-gun bullets, 
and some ten civilian employees of PAA 
had been killed.^^ The enemy did not lose 
a single bomber although "severar' were 
damaged by antiaircraft fire." The Ma- 
rine combat air patrol, well above the raid 
and momentarily scouting to the north, 
had not made contact. These pilots re- 
turned for landing shortly after the at- 
tack, and by a final stroke of ill fortune 
Captain Henry T. Elrod damaged his 
propeller seriously on a mass of bomb 

Wake's defenders were most concerned 
that this first raid had struck almost be- 
fore they knew that enemy planes were 
overhead. The rain squall had helped the 
Japanese, but the atoll's lack of early- 
warning equipment was almost as bene- 
ficial to the enemy. The garrison needed 
radar, but none was available. Through- 
out the siege the Japanese planes con- 
tinued to elude the most vigilant visual 
observation, and with the sound of their 
engines drow ned by the booming surf they 
would often have their bombs away before 
they were spotted. 

Damage control began at the airstrip as 
soon as the enemy departed. Casualties 
went to the one-story contractor's hos- 
pital which had been taken over as the 
island aid station," the dead were placed 

" Cunningham Interview, 5. 

''JICPOA Item No 4986, Professional note- 
book of Ens T. Nakamura, UN, 1941-1943, 
25Feb44, hereinafter cited as Nakamnra Note- 

" Tlie battalion surgeon of the 1st Def BnDet, 
Lt (jg) Gustave M. Kahn (MC), USN, was ably 

in a reefer box at Camp Two, and able- 
bodied aviation personnel turned their 
attention to the airplanes and to the gaso- 
line fires. The three planes still able to 
fly were sent up on combat air patrol. In 
the sky they would be safe from another 
surprise raid. Crews and officers re- 
organized and reallocated jobs. Second 
Lieutenant John F. Kinney became engi- 
neering officer to replace First Lieutenant 
Graves who had been killed.^^ Kinney's 
principal assistant was Technical Ser- 
geant William J. Hamilton, an enlisted 
pilot, and these two men began salvaging 
tools and parts from burned planes. 
Their efl^orts immeasurably aided future 
operations of VMF-211. Captain Her- 
bert C. Freuler reorganized the ordnance 
section. Lieutenant David D. Kliewer took 
over the radio section, and Captains Elrod 
and Tharin supervised construction of in- 
dividual foxholes, shelters, and infantry 
defensive works in the VMF-211 area. 
Other work included mining the airstrip 
at 150-foot intervals with heavy dynamite 
charges to guard against airborne land- 
ings. Furrows were bulldozed through- 
out the open ground where such landings 
might take place, and heavy engineering 
equipment was placed to obstruct the run- 
way at all times when friendly planes 
were not aloft. Plans called for continua- 
tion of the dawn and dusk reconnaissance 
flights, and for the initiation of a noon 
combat air patrol as well. It was hoped 
that these patrols could intercept subse- 
quent enemy raids. 

assisted by his civilian colleague, Dr. Lawton 
M. Shank, the contractor's surgeon, whose cool- 
ness and medical efficiency throughout the siege 
won high praise. 

^^ Kinney Interview, 4. 



Elsewhere on the atoll new defense 
work progressed just as rapidly.*® Em- 
placements, foxholes, and camouflage 
were improved at all battery positions. 
A Navy lighter loaded with dynamite sur- 
rounded by concrete blocks was anchored 
in Wilkes channel to guard this dredged 
waterway. Telephone lines were re- 
paired, key trunk lines were doubled 
wherever possible, and every possible at- 
tempt was made to bury the most impor- 
tant wires." Construction of more du- 
rable and permanent command posts and 
shelters began before the day ended in a 
cold drizzle. Working that night under 
blackout restrictions, aviation Marines 
and volunteer civilians completed eight 
blast-proof aircraft revetments. The 
atoll's four operational planes were thus 
relatively safe within these revetments 
when 9 December dawned bright and 
clear, and Captain Elrod's plane also, was 
in a bunker undergoing repairs to its pro- 
peller and engine. 

General quarters sounded at 0500, 45 
minutes before dawn, and the defense 
commander set Condition 1. This readi- 
ness condition required full manning 
of all phone circuits, weapons, fire con- 
trol instruments, and lookout stations. 
The four F4F-3's warmed up and then 

" Approximately ten per cent of the civilian 
workers volunteered for military or defensive 
duties, and some attempted to enlist. Many of 
these men served with heroism and eflBciency 
throughout the operation. 

" "Surface lines could not seem to stand up 
although they were all paralleled. We wanted to 
bury them, but we could not do so by hand . . . 
considering the scarcity of men to do the work. 
We could not obtain permission to use the ditch 
diggers of the contractors. . . ." LtCol C. A. 
Barninger reply to HistSec, HQMC question- 
naire, 18Feb47, R-0, hereinafter cited as Barn- 

took off at 0545 over Peacock Point. They 
rendezvoused in section over the field and 
then climbed upward to scout 60- to 80- 
mile sectors along the most probable routes 
of enemy approach. At 0700 the fighters 
finished their search without sighting any 
enemy planes and then turned back toward, 
the atoll. There the defense detachment 
shifted to Condition 2 which required that 
only half the guns be manned, and that 
fewer men stood by the fire control in- 
struments. This permitted Marines to get 
after other necessary work around their 
positions. At the airstrip Lieutenant Kin- 
ney continued work on Elrod's plane, and 
the squadron's engineering problem made 
it evident that hangar overhaul and black- 
out facilities had to be set up. Major Put- 
nam decided to enlarge two of his new 
plane shelters for this purpose. Entrance 
ramps were cut below ground level, and 
the revetments were roofed with "I" 
beams, lumber, and lightproof tarpaulins. 
These expedients allowed extensive over- 
haul and maintenance at all hours, and 
provided maximum protection for planes 
and mechanics. 

As the morning wore on, men began to 
work closer to their foxholes and to keep 
a wary eye skyward. A dawn takeoff 
from the nearby Japanese-mandated Mar- 
shalls could bring a second Japanese 
bomber raid over Wake at any time after 
1100. This "clock- watching" was justi- 
fied. Disgustingly prompt, enemy planes 
from the Twenty -Fourth Air Flotilla at 
Roi arrived at 1145.** Marine Gunner 
H. C. Borth spotted them first from the 
water tank OP, and he shouted the warn- 
ing over the "J"-line circuit. Seconds 
later the air-ground radio (again in opera- 
tion with makeshift equipment) passed 

Enemy Notes, 1. 



this alarm to the combat air patrol, and 
battery crewmen rushed to general quar- 
ters. Soon three bui-sts of antiaircraft 
fire, the new alarm signal, ^^ were explod- 
ing from all sectors, and Wake stood by 
for its second attack of the war. 

The leading Japanese planes ap- 
proached from the southeast at 13,000 
feet, and antiaircraft batteries on Peale 
Island and Peacock Point opened fire just 
before the first bombs were released. 
Minutes earlier the combat air patrol had 
made contact with one flank of the Japa- 
nese planes south of Wake, and Lieu- 
tenant Kliewer and Technical Sergeant 
Hamilton managed to cut oif a straggler. 
They shot it down despite hot return fire 
from a top turret, and as the enemy plane 
spun away in flames the ground batteries' 
3-inch shells began to burst among the 
Japanese. The Marine fighters broke con- 
tact and withdrew. 

The first sticks of bombs exploded 
around Batteries E and A on Peacock 
Point and damaged a 3-inch gun in the E 
Battery position and a range finder at 
Battery A. Other bombs crashed along 
the east leg of Wake Island and into 
Camp Two. There direct hits destroyed 
the hospital, the civilian and Navy bar- 
racks buildings, the garage, blacksmith 
shop, a storehouse, and a machine shop. 
The falling bombs then straddled the 
channel at this tip of Wake and began to 
rain down on Peale Island. They made 
a shambles of the Naval Air Station 
which was still under construction, and 

scored a direct hit on the radio station. 
This destroyed most of the Navy's radio 
gear.^" Meanwhile the antiaircraft guns 
continued to fire into the tight Japanese 
formation, and five bombers were smok- 
ing by the time Peale Island was hit. A 
moment later one of these planes burst 
into flames and blew up in the air. That 
was Wake's second certain kill. The 
others limped away still smoking.^^ 

The hospital burned to the ground 
while the two surgeons saved first the 
patients and then as much medical sup- 
plies and equipment as they had time to 
salvage. Camp Two and the Naval Air 
Station were now as badly wrecked as the 
aviation area had been on the previous 
day, and four Marines and 55 civilians 
had been killed. But the defenders had 
learned some lessons, and the Japanese 
were not to have such an easy time here- 
after. Major Putnam summed it up: 

The original raid . . . was tactically well con- 
ceived and skillfully executed, but thereafter 
their tactics were stupid, and the best that can 
be said of their skill is that they had excellent 
flight discipline. The hour and altitude of their 
arrival over the island was almost constant and 
their method of attack invariable, so that it was 
a simple matter to meet them, and they never, 
after that first day, got through unopposed. . . .^ 

Defenders spent that afternoon collecting 
wounded, salvaging useful items from 
blasted ruins, and moving undamaged in- 
stallations to safer spots. These jobs were 
to become painfully familiar on succeed- 
ing afternoons. 

"Wake did not have an air raid alarm, and 
this traditional three-shot signal was the only 
alternative. Defenders tried to make an alarm 
system with dismounted auto horns wired to 
storage batteries, but it never worked. Last Man 
Off Wake Island, 65, 122. 

""CO NAS Wake Rpt to ComFourteen, 20- 
Dec41, 1-2. 

^ A Japanese report indicates that 14 of these 
bombers were damaged by antiacraft fire during 
this attack. Nakamura Notebook. 

^ Putnam Rept, 10. 



The Japanese attack on Battery E at 
Peacock Point and along the island's east 
leg suggested to Major Devereux that 
the enemy would plan their raids in a 
logical sequence to pass over the atoll's 
long axis. On the previous day they had 
struck Wake's aviation, and now they had 
bombed not only the Naval Air Station 
but the 3-inch battery which had engaged 
them so promptly during that first raid. 
Thus Peacock Point was particularly vul- 
nerable, and to protect his remaining 
antiaircraft weapons, Devereux ordered 
Battery E to shift to a new site some six 
hundred yards east and north. There the 
battery could manage its job equally well. 
And to make sure that its fire power did 
not suffer, the battery drew one of the 
unused 3-inch guns assigned to the "phan- 
tom" Battery F on Wilkes. This weapon 
replaced the one damaged by bombs. 

To provide new hospital facilities, am- 
munition was cleared from the two 
most widely-separated reinforced con- 
crete magazine igloos, and these were con- 
verted into underground medical centers. 
Each measured 20 by 40 feet and could 
accommodate 21 hospital cots. They met 
blackout requirements, and with lights 
furnished by two small generators could 
be operated efficiently at night. Medical 
supplies were divided between the two 
aid stations. Dr. Kahn was in charge of 
the Marine hospital in the southern shel- 
ter, and Dr. Shank maintained the Navy- 
civilian facility at the north end of the 
row of magazine igloos. Both were in use 
by nightfall that day. 

During the night Battery E displaced 
to its new position. Aided by contractor's 
trucks and almost 100 civilian volunteers, 
Marines moved the guns, sandbags (too 
valuable and scarce to be left behind), 
fire control equipment, and ammunition. 

Emplacements were dug at the new site, 
sandbags refilled, and the guns readied for 
action. By 0500, just in time for dawn 
general quarters, the battery was in posi- 
tion and ready to fire.^^ Dummy guns 
were set up at the old position. 

On 10 December the Japanese con- 
firmed Devereux's theory that they would 
maintain certain patterns of approach 
and attack. At about 1045, 26 enemy 
bombers appeared, this time from the east. 
Again VMF-211 intercepted, and some of 
the bombers were hit before they reached 
the atoll. Captain Elrod, leading the 
fighters, shot down two enemy planes 
after the 3 -inch guns began to fire. 
Bombs hit Battery E's abandoned posi- 
tion at Peacock Point, but the new site 
was not threatened. On Peale Island 
Battery D received two successive passes 
by one enemy flight division. The first 
pass scored a damaging hit on the bat- 
tery's powerplant, but the guns continued 
to fire on barrage data. One plane burst 
into flames. 

On Wilkes Island, undamaged from the 
earlier raids, one stick of bombs lit 
squarely on a construction dump where 
125 tons of dynamite were cached west of 
the "New Channel." " The resultant ex- 
plosion stripped most of the underbrush 
off Wilkes, detonated all 5- and 3-inch 
ready ammunition at battery positions,^^ 

" LtCol W. W. Lewis reply to HistSec, HQMC 
questionnaire, 28Feb47, 1, hereinafter cited as 

" The "New Channel" was a partially-com- 
pleted waterway through the center of Wilkes. 

" By this time the Btry F position was being 
activated, but it was not as yet in full commis- 
sion. Marine Gunner McKinstry, with naval per- 
sonnel and volunteer civilians, had started that 
morning to form an antiboat battery with this 
unit's three guns and the damaged gun inherited 
from Btry E. 



and swept Battery L's emplacement clean 
of accessories, light fittings and other 
movable objects. Fortunately only one 
Marine was killed. Four others were 
wounded, and one civilian sustained 
shock. But materiel-wise. Battery L was 
in serious shape. All fire control instru- 
ments except the telescopes on Gun 2 had 
been blasted away or damaged beyond re- 
pair, the gun tubes were dented, firing 
locks were torn off, and traversing and 
elevating racks were burred and distorted. 
Equipment loss at Battery F, organizing 
that morning, was less serious. One gun 
was damaged from blast and flying de- 
bris. In addition, the 60-inch searchlight 
on Wilkes had been knocked end over end. 
This seriously damaged the light's deli- 
cate arcs, bearings, and electronic fittings. 
After this raid Major Devereux again 
ordered Battery E to displace. This time 
it would set up north of the airstrip and 
near the lagoon in the crotch of Wake. 
The dummy guns at Peacock Point, dam- 
aged by this third raid, were refurbished 
during the afternoon of 10 December, and 
Battery E's unmanned fourth gun was de- 
tached for antiboat emplacement else- 
where.^" Battery E's new position .would 
be most advantageous, the battery com- 
mander reasoned : 

Most all bombing runs were made from the 
east or west and the bombs were dropped along 
the length of the island. In this position the 
Japanese must make a run for the battery alone 
and most of the bombs would be lost in the 

That night the battery personnel sweated 
through their second displacement, and 

by next morning they were in position 
and again ready to shoot. 


After the Pearl Harbor attack. Presi- 
dent Roosevelt warned the American peo- 
ple to be prepared for the fall of Wake. 
Yet before the Arizona's hulk stopped 
burning, plans were underway to send re- 
lief to the atoll. But with much of the 
Pacific Fleet on the bottom of Pearl Har- 
bor, little assistance could be provided. 
Wake, like other outer islands, would 
stand or fall on its own unless it could be 
augmented from the meager resources 
then at Pearl Harbor. Marine forces on 
Oahu included two defense battalions, the 
3d and 4th,^® elements of the 1st Defense 
Battalion, and miscellaneous barracks and 
ships' detachments. Any personnel sent 
to relieve Wake would have to come from 
these units, and that meant that other im- 
portant jobs would have to be slighted. 
There was a limited source of equipment 
including radar and other supplies at 
Pearl Harbor in the hands of the Marine 
Defense Force quartermaster; and fighter 

^^ This 3-inch gun, which figured conspicuously 
in the later defense, was located south of the 
airstrip and the VMF-211 area. 

" Lewis, 2. 

^ Unless otherwise noted the material concern- 
ing the Relief Expedition is derived from a mag- 
azine article by LtCol R. D. Heinl, Jr., "We're 
Headed for Wake," MC Gazette, June 1^6. 

** This battalion, which during 1941-1943 ex- 
ecuted more overseas displacements than any 
other defense battalion in the Fleet Marine 
Force, pulled out of Guantanamo Bay, Cuba, 
during late October 1941, moved secretly through 
the Panama Canal, and arrived at Pearl Harbor 
on Monday 1 December. On 7 December the bat- 
talion manned a 3-inch battery at the Navy Yard, 
and also served some antiaircraft machine guns. 
Since it had just completed this oversea move- 
ment, and had its equipment ready for service, 
the 4th was a logical choice for its eventual 
role in the attempt to relieve Wake. 



aircraft, needed almost as much as radar, 
were already en route from San Diego on 
board the USS Saratoga?^ 

On 9 December ^^ Admiral Kimmel's 
staff decided to send relief to Wake in a 
task force built around this carrier, Cruis- 
er Division 6 (cruisers Astot'ia, Minneapo- 
lis, and /San Francisco), the nine destroy- 
ers of Destroyer Squadron 4, the seaplane 
tender Tangier, which would carry troops 
and equipment, and the fleet oiler Neches. 
These ships would comprise Task Force 
14. Wliile it sailed for Wake, Task Force 
11 built around the USS Lexington, would 
make divereionary strikes in the vicinity 
of Jaluit some 800 miles south of Wake. 
A third task force, commanded by Vice 
Admiral Halsey in the carrier Enterprise, 
would provide general support by con- 
ducting operations west of Johnston Is- 

Men and equipment to aid Wake would 
be drawn from the 4th Defense Battalion, 
and on 10 December this unit was alerted 
for immediate embarkation. The destina- 
tion was not announced, but it did not re- 
quire much imagination for rumor to cut 
through military secrecy. "We're going 
to Wake" was the word that circulated all 
day while the batteries prepared to mount 
out. By nightfall the personnel and 
equipment were squared away, and units 
groped about in the blackout to assemble 
their gear for loading. But in the midst 
of this work came orders to knock off and 

^ These planes comprised VMF-221. The Sar- 
atoga had departed at maximum speed on 8 De- 
cember (9 December on Wake). Ship's Log USS 
Saratoga, December 1941, hereinafter cited as 
Saratoga log. 

'^ Throughout this section dealing with the re- 
lief attempt, west longitude dates and local times 
are used. 

'' CinCPac OPlan 39-41, 15Dec41, 2. 

return to original battery positions. The 
CinCPac staff wanted to make a complete 
new study of the Pacific situation before 
it sent this relief off to Wake.^^ Besides, 
the task force had to await the arrival of 
the Saratoga. 

CinCPac finally decided to make the 
attempt to reinforce Wake, and em- 
barkation of certain units of the 4th De- 
fense Battalion began two days later, on 
12 December. By this time the Wake de- 
fenders had sent a partial list of their most 
critical needs, and Pearl Harbor supply 
activities filled this as best they could. 
I'hese important items, which were loaded 
in the Tangier at pier 10 in the Navy 
Yard, =** included an SCE-270 early-warn- 
ing radar unit and an SCR-268 radar set 
for fire control. Also stowed on board 
were 9,000 rounds of 5-inch ammunition, 
12,000 of the 3-inch shells with 30-second 
time fuzes, more than three million rounds 
of belted ammunition for .50 and .30 cali- 
ber machine guns, quantities of grenades, 
ammunition for small arms, barbed wire, 
antipersonnel mines, and additional engi- 
neering tools. Other equipment would en- 
able the men at Wake to repair their 
bomb-damaged weapons. This included 
three complete fire control and data trans- 
mission systems for 3-inch batteries, 
needed replacement equipment for the 
atoll's 5-inch guns, electrical cable, ord- 
nance tools, and spare parts. 

Units of the 4th Defense Battalion em- 
barked for this expedition included Bat- 
tery F with 3-inch guns, Battery B with 
5-inch guns, a provisional machine gun de- 
tachment drawn from Batteries H and I, 

** Notes of interview by Capt S. E. Morison, 
USNR, with RAdm C. H. McMorris, 13Jan47, 
hereinafter cited as McMorris Interview. 

'* USS Tangier log, December 1941. 



and a headquarters section drawn from 
the Headquarters and Service Battery of 
the defense battalion. First Lieutenant 
Robert D. Heinl, Jr., commanded tliis 
force when it completed embarkation on 
13 December, but the command passed to 
Colonel H. S. Fassett just prior to the de- 
parture of the task force two days later.^^ 
After loading, the Tangier moved to the 
upper harbor ^ where Rear Admiral 
Fletcher's Cruiser Division 6 Avaited for 
the Saratoga. The carrier came in to fuel 
on the 15th, " and the task force sortied 
late that day and set course for Wake. 

8-11 DEC EMBER ^^ 

Admiral Inouye, commanding the Japa- 
nese Fourth Fleet at Truk, had set numer- 
ous projects and operations in motion on 
8 December. Current war plans called 
for him to capture and develop Wake, 
Guam, and certain Gilbert islands includ- 
ing Makin and Tarawa. By 10 Decem- 
ber, when Guam fell, Inouye could check 
off all these jobs except the one at Wake.^^ 
Despite its small size this atoll was giving 
the admiral and his people at Truk and 
Kwajalein some moments of worry. Tlie 
other islands had fallen to them with little 

^ Col Fassett was to become Island Com- 
mander at Wake when the relief force arrived 
at the atoll. ConiFourteen orders to Col Fassett, 

*■ Tangier log, op. cit. 

^' The Saratoga arrived in the Hawaiian area 
during the night of 14 December, but she could 
not enter the harbor until next day when the 
antisubmarine nets were opened. Saratoga log. 

"^ Except as otherwise noted material in this 
section dealing with Japanese operations is de- 
rived from Capture of Wake; Wake Attack; 
Nakamura Notebook; Enemy Notes. 

'" Campaigns of the Pacific War, 47. 

trouble, but they knew that Wake's de- 
fense was in better shape. Tliey estimated 
that this atoll was defended by about 
1,000 troops and 600 laborers. Wake's 
fighter planes were aggressive, and the 
flak from the island was at least prompt 
and determined. Between the Marine 
planes and this flak the Twenty-Fourth 
Air Flotilla surely had lost five of its 
planes, not counting four more "smokers" 
that the Wake defenders fervently lioped 
never made it back to Roi. 

This Tu^enty -Fourth Air Flotilla was 
composed of Air Attack Forces One and 
Three. Force One flew shore-based 
bombers, and Force Three operated 
approximately 15 four-engined patrol 
bombers (probably Kawanishi 97s). 
Force One based on Roi, while Force 
Three^ which was also bombing or scout- 
ing Baker, Howland, Nauru, and Ocean 
Islands, flew out of Majuro Atoll 840 
miles south of Wake. The commander 
of this air flotilla had the mission of 
softening Wake for capture, and he 
was going about it in a creditable fasli- 
ion. First he struck the airstrip to clear 
out the fighter planes, and then he figured 
to come back with the sky to himself and 
finish off his job. Subsequent targets had 
been the Naval Air Station, seaplane fa- 
cilities, and other installations. With 
these missions accomplished, the pilots of 
the Twenty -Fourth Air Flotilla could 
settle down to the methodical business of 
taking out the antiaircraft and seacoast 
batteries. Thus the raid of 10 December 
concentrated on Peale where poor bomb- 
ing and Battery D's fire held the Japanese 
to no gains, and on Wilkes where bombs 
set off' the dynamite cache. 

After those three strikes the Japanese 
decided Wake was ripe for a landing, and 



the job went to Rear Admiral Kajioka 
who commanded Destroyer Squadron 6 
in his new light cruiser Yubari. Kajioka 
planned to land 150 men on Wilkes Island 
to control the dredged channel, and 300 
men on the south coast of Wake Island to 
capture the airfield. An alternate plan 
called for landings on the north and 
northeast coasts, but the admiral hoped 
to avoid these beaches unless unfavorable 
winds kept his men away from the south 
side of the atoll. The Japanese expected 
that a landing force of only 450 men 
would face a difficult battle at Wake, but 
this force was the largest that Admiral 
Kajioka could muster at this early date in 
the war. But if things hit a snag, de- 
stroyer crews could be used to help storm 
the beaches. The naval force at Admiral 
Kajioka's disposal included one light 
cruiser (the flagship), two obsolescent 
light cruisers for fire support and cover- 
ing duties, six destroyers, two destroyer- 
transports, two new transports, and two 
submarines.*" The Tioenty -Fourth Air 
Flotilla would act as his air support. 
Wake was so small that the admiral did 
not consider carrier air necessary. 

The 450 men of the landing force con- 
stituted Kajioka's share of the special 
naval landing force personnel assigned 
to the Fourth Fleet. It is probable that 
they were armed with the weapons typical 
to a Japanese infantry unit of company 
or battalion size, and that their weapons 

*" Yubari; Tatsuta and Tenryu (2 old light 
cruisers, comprising Cruiser Division 18) ; Oite, 
Hayate, Mutsuki, Kisaragi, Mochizulci and 
Yayoi (6 older destroyers, comprising Destroyer 
Division 29 and 30) ; Patrol Boats 32 and 33, 
so-called (actually old destroyers converted into 
light troop-carrying craft with missions similar 
to the American APD) ; and Kongo Maru and 
Konryu Maru, both medium transports. 

included light machine guns, grenade 
launchers, and possibly small infantry can- 
non. It is likely that assault troops were 
embarked in the two old destroyer-trans- 
ports (Patrol Craft 32 and 33), while 
the garrison and base development eche- 
lon was assigned to the medium-size trans- 
ports. The assault shipping from Truk 
arrived at Roi on 3 Dexiember, and on 9 
December *^ the force sortied on a circui- 
tous route for Wake. 

The Japanese expected no American 
surface opposition, but they nonetheless 
screened their approach with customary 
caution. Two submarines scouted 75 
miles ahead of the main body, and these 
boats were to reconnoiter Wake prior to 
the arrival of the task force.*^ Specifi- 
cally they would try to find out whether 
the atoll defenders had any motor torpedo 
boats. Behind these submarines, and 10 
miles forward of the main body, a picket 
destroyer maintained station from which 
it would make landfall and conduct a fur- 
ther reconnaissance. Ships of the task 
force neared Wake on the evening of 10 
December. The weather was bad with 
high winds and heavy seas, but there was 
advantage even in this. The squalls pro- 
vided a natural screen behind which the 
approach would surely remain unde- 
tected. Reports from the submarines and 
the screening destroyer indicated that 
Wake was not aware of the Japanese £^p- 
proach, and at 0300 on 11 December the 

" Capture of Wake, II. 373 lists this date as 8 
December, but other dates from this authority 
are consistently one day behind, and it is there- 
fore probable that the date of 9 December is 

" The submarines were scheduled to arrive at 
Wake prior to dawn, and it is therefore not 
clear how they expected to make a visual re- 
connaissance that would be of much value. 



task force made landfall and prepared to 
disembark the landing force. From Kaji- 
oka's flagship Wake was barely visible 
while the admiral led his force to bom- 
bardment and debarkation stations five or 
six miles off the atoll's south shore. 


In spite of Wake's black silent appear- 
ance to the Japanese, the atoll defenders 
had spotted the enemy. Lookouts re- 
ported ships in sight just prior to 0300, 
and as the shadowy outlines drew closer 
Devereux decided they formed an enemy 
force which included cruisers, destroyers, 
and some auxiliaries. The garrison went 
to general quarters, and Devereux ordered 
Major Putnam to delay the takeoff of his 
four airplanes until after the shore bat- 
teries began to fire. And these batteries 
were ordered to hold their fire *^ until they 
received orders to open up. Major Dev- 
ereux reasoned that the enemy force could 
outgun his defense force, and that prema- 
ture firing would only reveal the location 
and strength of the seacoast batt-eries and 
rob them of a chance to surprise the 

Meanwhile the enemy force was having 
trouble with the bad weather that had 
screened its approach. Assault troops 
found it difficult to make their transfer to 
sea-tossed landing craft, and some of these 
craft overturned or became swamped in 
the high waves. By dawn at 0500, the 
flagship Yubari, still in the van, reached a 

" Cdr Cunningham's postwar report states 
that Maj Devereux wanted to illuminate the 
enemy force with searchlights and to open fire 
much sooner, but that this request was denied. 
Cunningham Interview, 7. Devereaux denies 
this, and he is supported by virtually all other 
records of the action. 

position approximately 8,000 yards south 
of Peacock Point. There she turned west- 
ward and commenced a broadside run par- 
allel to the south shore of Wake. The 
other enemy ships followed generally 
along this course but kept approximately 
1,000 yards further to seaward. Al- 
though the Japanese were not aware of it, 
the Yubari was being tracked along this 
course by the 5-inch guns of Battery B on 
Peacock Point. The camouflage had been 
removed from battery positions so that 
the guns could train.^* 

A few minutes later, the Yubari and the 
other two cruisers {Tatsuta and Tenryu) 
opened fire at area targets along the south 
shore of Wake. These salvos laddered the 
island from Peacock Point to the vicinity 
of Camp One. The high-velocity 6-inch 
shells which hit near Camp One ignited 
the diesel-oil tanks between the camp and 
Wilkes Channel, and only a repetition of 
Devereux's order to hold fire restrained 
Lieutenants Clarence A. Barninger and 
John A. McAlister, respectively com- 
manding the 5-inch batteries at Peacock 
and Kuku Points, from returning fire. 
The other Japanese ships, following the 
cruiser and destroyer screen, maneuvered 
to take stations for their various missions. 

After completing her initial firing run 
the Yubari, apparently accompanied by 
the two destroyer-transports, reversed 
course in a turn which closed the range on 
Wake. By this time it was daylight, and 
by 0600 these ships were some 3,500 yards 
south of Battery A on Peacock Point.*^ 

" Barninger, 4. 

^ The range finders on the 5-inch guns of 
Btrys A and L had been rendered inoperative by 
previous bombings, and ranges therefore had to 
be estimated. This resulted in considerable 
variance among the later reports of this action. 
These discrepancies undoubtedly were aggra- 



Battery A, with no range finder, had esti- 
mated the range to these sliips, and the 
range section personnel were plotting the 
target while the gun section crews stood 
by to fire. The order to fire came from 
Devereux's command post at 0615, and the 
guns at Peacock Point opened fire on the 
Yubari and the ships with her while Bat- 
tery L engaged the other enemy ships 
within range of Wilkes Island. Battery 
A's first salvo went over the Japanese flag- 
ship, and Lieutenant Barninger ordered 
the range dropped 500 yards. This fire 
from the beach caused the cruiser to veer 
away on a zig zag course, and to concen- 
trate her fire on the Battery A guns. Her 
shots straddled the Marine positions as she 
pulled away rapidly. Barninger adjusted 
as best he could for the evasive tactics of 
the Japanese ship, and his guns soon 
scored two hits. Both shells entered the 
cruiser at the waterline amidships on her 
port side, and the ship belched steam and 
smoke as she slackened speed. Two more 
shells then caught her slightly aft of these 
first wounds, and she turned to starboard 
to hide in her own smoke. A "destroyer 
then attempted to lay smoke between the 
troubled cruiser and the shore battery, but 
it was chased away by a lucky hit from 
a shell aimed at the cruiser. The Yuhmi 
continued to fire at Peacock Point until 
her 6-inch guns could no longer reach the 
island. Then, listing to port, she limped 
smoking over the horizon.** 

Meanwhile Battery L had opened up 
from Wilkes on the three destroyers, two 

vated by the long dispersion pattern characteris- 
tic of these flat trajectory naval weapons. 

*" The Btry A commander, whose comments are 
the source of this account of action against Adm 
Kajioka's flagship, believes that his guns scored 
two more hits on the cruiser before she got out of 
range. Barninger, 4^5. 

transports, and the light cruisers Tatstita 
and Tenryu which had broken off from the 
Yubari at the west end of her first firing 
run. These cruisers and transports 
steamed north at a range of about 9,000 
yards southwest of Kuku Point while the 
destroyers (probably Destroyer Division 
29 consisting of the Hayate, Oite, and 
either the Mutsvki or Mochizuki)*'' 
headed directly for shore and opened fire. 
At about 4,000 yards from the island they 
executed a left (westward) turn, and the 
Hayate led them in a run close along the 
shore. At that point Battery L opened 
fire. At 0652, just after the third two- 
gun salvo, the Hayate erupted in a violent 
explosion, and as the smoke and spray 
drifted clear, the gunners on Wilkes could 
see that she had broken in two and was 
sinking rapidly. Within two minutes, at 
0652, she had disappeared from sight.** 
This prompted such spontaneous celebra- 
tions in the Battery L positions that a 
veteran noncommissioned officer had to re- 
mind the gun crews that other targets re- 

Fire then shifted to Oite, next in line 
behind the Hayate. This destroyer was 
now so close to shore that Major Devereux 
had difficulty restraining his .30 caliber 
machine gun crews from firing at her. A 
5-inch gun scored one hit before the on- 
shore wind carried smoke in front of the 
target. With this concealment, the de- 
stroyers turned to seaward away from 
Battery L. Marines fired several more 
salvos into the smoke, but they could not 
spot the splashes. Some observers on 
Wilkes thought they saw the Oite transfer 

" Wake Attack. 

** Piatt, op cit., 3. The Hayate thus was the 
first Japanese surface craft sunk during the war 
by U. S. naval forces. 



survivors and sink, but reliable enemy rec- 
ords indicate only that she sustained 

Battery L now shifted fire to the trans- 
ports Kongo Maru and Konryu Maru 
then steaming approximately 10,000 yards 
south of Wilkes. One shell hit the lead- 
ing transport, and this ship also turned to 
seaward and retired behind a smoke 
screen which probably was provided by 
the two fleeing destroyers. Their course 
carried them past the transport area. By 
this time civilians on Wilkes had joined 
the defensive efforts as volunteer ammuni- 
tion handlers, and the battery next en- 
gaged a cruiser steaming northward 9,000 
yards off the west end of the island. This 
was either the Tenryu or the Tatsuta; but 
whatever her identity, she hurried away 
trailing smoke after one shell struck her 
near the stern. The departure of this 
ship, at about 0710, removed the last tar- 
get from the range of Battery L. In a 
busy hour, this unit had fired 120 5-inch 
shells which sank one destroyer, damaged 
another, and inflicted damage to a trans- 
port and a light cruiser. Two Marines 
had sustained slight wounds. 

Meanwhile the other half of the Jap- 
anese destroyer force {Destroyer Division 
30) ran into its share of trouble as it 
moved west of Kuku Point on a north- 
westerly course. Led (probably) by the 
Yayoi, these three destroyers at 0600 
steamed within range of Battery B's 
5-inch guns on Peale. The Marines 
opened fire on the leading ship, and the 
Japanese promptly raked Peale with re- 
turn salvos which scored hits in and about 
the positions of Batteries B and D. This 
shelling destroyed communications be- 
tween Battery B's guns and the battery 

command post, and put Gun Two out of 
action with a disabled recoil cylinder. 
Lieutenant Woodrow W. Kessler, the bat- 
tery commander, continued his duel with 
only one gun, and used personnel from 
Gun Two to help keep up the fire. Ten 
rounds later a shell caught the Yayoi in 
her stern and set her afire. Kessler then 
shifted his fire to the second ship which 
was maneuvering to lay a smoke screen 
for the injured Yayoi. Under this con- 
cealment all three destroyers reversed 
course and retired southward out of 

The Japanese force was now in full 
retirement. At 0700 Admiral Kajioka or- 
dered a withdrawal to Kwajalein. Bad 
weather and accurate Marine fire had 
completely wrecked the admiral's plan to 
take Wake with 450 men. But command- 
ers on the atoll took immediate precau- 
tions to guard against a dangerous relax- 
ation of defenses. They reasoned that the 
Japanese might have carrier aircraft 
ready to continue the attack which the 
ships had started, and Major Putnam was 
already aloft with Captains Elrod, Freu- 
ler, and Tharin to reconnoiter the area 
from 12,000 feet. When this search lo- 
cated no enemy aircraft or carriers, the 
Marine pilots turned southwest to over- 
take the retiring Japanese task force. 
The fliers found the enemy little more 
than an hour's sail from Wake, and they 
swept down to attack. 

Captains Elrod and Tharin strafed and 
bombed two ships (probably the cruisers 
Tenryu and Tatsuta),^" and got their 
planes damaged by heavy antiaircraft fire 

' Enemy Notes, 1. 

^ The VMF pilots were not sure about the 
identification of their targets, but a consultation 
of all available sources of information seems to 
substantiate this account of the action. 



from these two targets. But the Tenryu 
suffered bomb damage to her torpedo bat- 
tery, and the Tatsuta's topside radio shack 
was hit. Captain Freuler landed a 100- 
pound bomb on the stern of the transport 
Kongo Maru^ and saw his target flare up 
with gasoline fires. After dropping their 
two bombs each, the fliers hurried back to 
Wake to rearm. 

Two fresh .pilots, Lieutenant Kinney 
and Technical Sergeant Hamilton, substi- 
tuted for two of the original fliers during 
one of these shuttles between the atoll and 
the enemy ships, and the air attacks con- 
tinued for a total of 10 sorties during 
which the Marines dropped 20 bombs and 
fired approximately 20,000 rounds of .50 
caliber ammunition.^' The destroyer 
Kisaragi^ probably hit earlier by Captain 
Elrod, finally blew up just as Lieutenant 
Kinney nosed over at her in an attack of 
his own. One of the destroyer-transports 
also sustained damage from the air 

This action w^as not all "ducks in a bar- 
rel" to the Marine fliers, and any damage 
to the scanty Wake air force was a serious 
one. Japanese flak cut the main fuel line 
in Elrod's Grumman, and although he 
managed to get back to the atoll he demol- 
ished his plane in a crash landing amid 
the boulders along Wake's south beach. 
Antiaircraft fire pierced the oil cooler and 
one cylinder in Captain Freuler's plane. 
He returned to the field safely, but he fin- 
ished his approach on a glide with a dead 
engine that could never be repaired. 

Accurate assessment of enemy losses in 
this first landing attempt is not possible. 
Japanese records indicate, however, that 

the destroyer Hayate was sunk by shore 
batteries and the destroyer Kisaragi by 
the VMF-211 bombs. Two more destroy- 
ers, the Oite and the Yayoi, were damaged 
as was a destroyer-transport. The trans- 
port Kongo Maru was bombed and set 
afire. All three cruisers ( Yuhari, Tatswta^ 
and Ten7'yu) received injuries from air or 
surface attacks.^^ 

Japanese personnel casualties can be 
fixed only approximately. Assuming that 
the two sunken destroyers w^ere manned 
by crews comparable to those required by 
similar U. S. types (about 250 officers and 
men per ship) , it would be logical to claim 
approximately 500 for these two losses 
with the fair assumption that few if any 
survivors escaped in either case. Person- 
nel losses on the other seven ships dam- 
aged are not known, but it must be 
assumed that casualties did occur.^^ 

' CO VMF-211 Rept to CO MAG-21, 20Dec41. 

" The widely-credited claim, originated in 
good faith, that dive-bombing attacks sank a 
cruiser off Wake cannot be supported. All three 
cruisers returned to Wake less than two weeks 
later to support the final attack on the atoll. 
The officially established occasion of the loss 
of each is as follows: Yubari (Philippine Sea, 
27 Apr44) : Tenryu (Bismarck Sea, by submarine 
action, 18Dec42) ; Tatsuta (off Yokohama, by 
submarine action, 13May44). As indicated in 
the text the violent explosion and sinking of the 
Kisaragi, combined with recognition inexperi- 
ence, probably accounts for the cruiser claimed. 
ONI StatisticalSec, "Naval Losses of All Na- 
tions," (located at NHD), 5Feb46, Table VIIL 

"" In a letter dated 22Nov51, Capt Tashikazu 
Ohmae, leading Japanese naval student of WW 
II, puts Japanese losses for this phase of the 
Wake operation at "nearly 500." Ohmae letter 
cited in Robert Sherrod, History of Marine 
Corps Aviation in World War II (Washington: 
Combat Forces Press, 1952), 41, hereinafter 
cited as Marine Air History. 

Wake Under Siege ^ 


Scarcely had the VMF-211 planes re- 
turned to the field before it was time for 
Lieutenants Davidson and Kinney to fly 
the only two serviceable fighters on the 
early midday combat patrol. It was 
then nearly 1000, almost time for the 
Japanese bombers to arrive, and the 
Marines soon spotted 30 of these enemy 
planes coming out of the northwest 
at 18,000 feet. Davidson downed two of 
these aircraft, and Kinney turned a third 
homeward with smoke trailing behind it. 
Then the fliers pulled away as the enemy 
formation entered the range of the Wake 

This antiaircraft fire splashed one 
bomber in the water off Wilkes and 
damaged three others. Bombs hit close to 
Battery D on Peale, and others exploded 
on Wake. There were no Marine casual- 
ties, and damage was slight, but the pat- 
tern of the attack convinced Devereux that 
the Japanese had spotted the position of 
Battery D. As soon as the attack ended 
he ordered this unit to displace from the 
neck of Toki Point to the southeastern 
end of Peale. ^ Marines and civilians be- 

' Unless otherwise noted the material in Chap 
3 is derived from Devereux Rept ; Putnam Rept; 
(officer's name) Repts, especially IstLt J. F. 
Kinney Rept, Major W. L. J. Bayler Rept, and 
LtCol B. D. Godbold Rept; Devereux Story. 

'At 1700, just prior to Battery D's dis- 
placement, a smoke bomb and a chain-flare 
of three red balls was sighted about two miles 
northeast of Toki Point. This signal was re- 

gan this displacement after dark. Sand- 
bags at the old position could not be re- 
claimed, and cement bags and empty am- 
munition boxes had to serve this purpose 
at the new location. The work was 
finished by 0445, and Battery D again 
was ready to fire. 

On 12 December the Japanese came to 
work early. Two four-engine Kawanishi 
patrol bombers arrived from Majuro at 
about 0500 and bombed and strafed Wake 
and Peale Islands. Bombs hit the airstrip 
but caused little damage. Captain Tharin, 
who had just taken off on the morning 
reconnaissance patrol, intercepted one of 
the big flying boats and shot it down. 
After this raid the Wake defenders went 
on with their work. Beach defenses were 
improved on Wilkes, and the ordnance of- 
ficer. Gunner Harold C. Borth, serviced 
Battery L's battered 5-inch guns. At the 
airfield Lieutenant Kinney managed to 
patch up one of VMF-221's cripples, and 
this brought the strength of the Wake air 
force up to three planes. Such work con- 
tinued for the remainder of the day. To 
the surprise of everyone on the atoll, the 

peated twice in the next 20 minutes. The sig- 
nificance of these signals has not been estab- 
lished. Japanese submarines were operating in 
the vicinity of Wake, however and it may be 
that the signals had something to do with rescue 
operations in which these boats were trying to 
aid survivors from a bombing raid or from the 
surface action. 




enemy did not arrive for the usual noon 

This freedom from attack was a wel- 
come and profitable interlude for the gar- 
rison. Captain Freuler, who had been at- 
tempting since the opening of the war to 
devise some means of employing welder's 
oxygen to augment the dwindling supply 
for the fighter pilots, finally managed, at 
great personal hazard, to transfer the gas 
from commercial cylinders to the fliers' 
oxygen bottles. Without this new supply 
the pilots could not have flown many more 
high altitude missions. 

Another important experiment failed. 
Marines tried to fashion a workable air- 
craft sound locator out of lumber. It was 
"a crude pyramidal box with four un- 
curved plywood sides," by Major Deve- 
reux's description. It was too crude to 
be of any value ; it served only to magnify 
the roar of the surf. 

That evening Lieutenants Kinney and 
Kliewer and Technical Sergeant Hamilton 
readied Wake's three planes for the final 
patrol of the day. Kliewer drew a plane 
that was always difficult to start, * and his 
takeoff was delayed for nearly fifteen 
minutes. While he was climbing to over- 
take the other fliers he spotted an enemy 
submarine on the surface some 25 miles 
southwest of the atoll. He climbed to 
10,000 feet and maneuvered to attack with 

' From 12 December until about 20 December, 
another day on which the enemy did not raid 
Walie, the recollections of surviving defenders 
sometimes are confused ■ beyond any possible 
reconciliation. This condition is acute for the 
period of 12-14 December inclusive. Sources 
reconstructing the events of those dates arrive 
at few compatible accounts. This volume at- 
temps to draw the best possible compromise from 
these conflicts. 

* Last Man off Wake Island, 120. 

the sun behind him. He strafed the Jap- 
anese boat with his .50 caliber guns, and 
then dropped his two bombs as he pulled 
out of his glide. Neither bomb hit, but 
Kliewer estimated that they exploded 
within fifteen feet of the target. Bomb 
fragments punctured his wings and tail 
as he made his low pull-out, and while he 
climbed to cruising altitude he saw the 
enemy craft submerge in the midst of a 
large oil slick.^ 

After their various activities on 12 De- 
cember, the atoll defenders ended the day 
with a solemn ceremony. A large grave 
had been dug approximately 100 yards 
southwest of the Marine aid station, and in 
this the dead received a common burial 
while a lay preacher from the contractor's 
crew read simple prayers. 

Next day the Japanese did not bother 
Wake at all, and Marine officers thought 
it possible that Kliewer's attack on the 
enemy submarine had brought them this 
day of freedom. The tiny atoll, fre- 
quently concealed by clouds, was a diffi- 

^ The fate of this submarine is not known. 
Enemy records are not clear. But after the fall 
of Walie, Kinney and other pilots were ques- 
tioned by a Japanese officer who asked them if 
they knew anything about a Japanese submarine 
that had disappeared in the vicinity of the atoll. 
This led Kliewer to believe that concussion from 
his bombs had finished off the submarine. The 
Japanese list two of their submarines (RO-66 
and RO-62) sunk 25 miles southwest of Wake 
on 17 Dec 1941. RO-66 was lost not to enemy 
action but to disaster, the Japanese said. The 
cause of the loss of RO-62 is not known, and it 
therefore may be assumed that there was some 
confusion as to the date. And since the Wake 
Marines had trouble remembering exact dates 
in their postwar reconstruction of specific events, 
it may be that Kliewer's bombs sank the RO- 
62. MilHistSec, SS, GHQ, FEC, Japanese Stud- 
ies in WWII No. 116, The UN in WWII, Feb- 
ruary 1952 (located at OCMH). 



cult place to find, and the Marines rea- 
soned that the Japanese were using sub- 
marine radios as navigational homing 
aids. Accurate celestial navigation would 
have been possible at night, but the bomb- 
ers had been making daylight runs of 
from 500 to 600 miles over water with no 
landmarks. By dead reckoning alone this 
would have been most difficult, yet the 
bombers hit at about the same time each 
day. This convinced some of the Marines, 
including Lieutenant Kinney, that the 
submarine had been leading them in.® 

But even this quiet day ' did not pass 
without loss. While taking off for the 
evening patrol, Captain Freuler's plane 
swerved toward a group of workmen and 
a large crane beside the runway. To 
avoid hitting the men or this crane, Freu- 
ler made a steep bank to the left. The 
plane lost lift and settled into the brush, 
a permanent wash-out. It was set up in 
the bone yard with other wrecks which 
were parked to draw bombs. 

December 14 started explosively at 0330 
when three four-engine Kawanishi 97 fly- 
ing boats droned over from Wotje ® and 
dropped bombs near the airstrip. They 
caused no dairiage, and the garrison made 
no attempt to return fire. But later that 
day the pilots of the Txoenty -Fourth Air 
Flotilla resumed their bombing schedule. 
Thirty shore-based bombers arrived from 
Roi at 1100, and struck Camp One, the la- 
goon off Peale, and the west end of the 
airstrip. Two Marines from VMF-211 

"Kinney Interview, 4. 

' On this date hot rations coolted in the con- 
tractor's galley were delivered to all battery po- 
sitions by a "chuck wagon." This service con- 
tinued for as long as possible thereafter. It was 
one of Mr. Teters' many contributions to the de- 

^ Enemy Notes, 1. 

were killed and one wounded, and a direct 
bomb hit in an airplane revetment fin- 
ished off another fighter plane, leaving the 
atolFs aviation unit only one plane that 
could fly.* Lieutenant Kinney, VMF- 
211's engineering officer, sprinted for the 
revetment where he was joined by Tech- 
nical Sergeant Hamilton and Aviation 
Machinist's Mate First Class James F. 
Hesson, USN,^" his two assistants. De- 
spite the fire which engulfed the rear end 
of the plane, these men accomplished the 
unbelievable feat of removing the undam- 
aged engine from the fuselage and drag- 
ging it clear. 

During his morning patrol flight of 15 
December, Major Putnam sighted another 
submarine southwest of Wake. But it ap- 
peared to have orange markings, and Put- 
nam did not attack. He thought it might 
be a Netherlands boat because he had ob- 
served markings of that color on Dutch 
airplanes in Hawaii in late 1941. Put- 
nam's examination of the craft caused it 
to submerge, however, and Marines later 
took significant notice of this when the 
regular bombing raid did not arrive that 
day. This seemed to add credence to the 
theory that submarines were providing 
navigational "beams" for the bombers. 

But the Kawanishi flying boats kept the 
day from being completely free of Jap- 
anese harassment. Four to six of these 
four-motored planes came over at about 
1800, and one civilian was killed when the 
planes made a strafing run along the atoll. 

" Wake File. 

" Hesson was a Navy aviation instrument re- 
pairman sent over to VMF-211 from the Naval 
Air Station after the first attack had played 
such havoc among the fighter squadron's 
ground personnel. He turned out to be an out- 
standing general aviation maintenance man. 



Their bombing was less effective. They 
apparently tried to hit Battery D on Peale 
Island, but most of the bombs fell harm- 
lessly into the lagoon and the others 
caused no damage. 

Meanwhile defensive work continued 
during every daylight hour not inter- 
rupted by such bombing raids. Another 
aircraft was patched up, personnel shel- 
ters for all hands had been completed in 
the VMF-211 area, and at Peacock Point 
Battery A now had two deep under- 
ground shelters with rock cover three feet 
thick.^* And before nightfall on 15 De- 
cember the garrison completed its destruc- 
tion of classified documents. This secur- 
ity work began on 8 December when the 
Commandant of the 14th Naval District 
ordered Commander Cunningham to de- 
stroy reserve codes and ciphers at the Na- 
val Air Station," but codes remained in- 
tact in the VMF-211 area. Now Major 
Bayler and Captain Tharin shredded 
these classified papers into an oil drum 
and burned them in a gasoline fire." 

On the 16th the Japanese made another 
daylight raid. Twenty-three bombers 
from Roi came out of the east at 18,000 
feet in an attempt to bomb Peale Island 
and Camp Two. Lieutenants Kinney and 
Kliewer, up on air patrol, warned the 
garrison of this approach, but the Marine 
fliers had no luck attacking the enemy 
planes. They did radio altitude informa- 
tion for the antiaircraft gunners, how- 
ever, and the 3-inch batteries knocked 
down one bomber and damaged four 
others. The Japanese spilled their bombs 
into the waters of the lagooon and turned 
for home. 

" Wake File. 

" Cunningham Interview, 4. 

" Last Man Off Wake Island, 112. 

But experience had taught the atoll de- 
fenders not to expect a rest after this day- 
light raid was over. The flying boats had 
become almost as persistent as the shore- 
based bombers, and at 1745 that afternoon 
one of the Kawanishis came down through 
a low ceiling to strafe Battery D on Peale 
Island. Poor visibility prevented the 
Marines from returning fire, but the at- 
tack had caused little damage. The plane 
dropped four heavy bombs, but these fell 
harmlessly into the lagoon. Marines who 
were keeping score — and most of them 
were — marked this down as Wake's 10th 
air raid. 

After this attack Wake had an uneasy 
night. It was black with a heavy drizzle, 
and maybe this put sentinels on edge just 
enough to cause them to "see things" — al- 
though no one could blame them for this. 
At any rate lookouts on Wilkes passed an 
alarm at 0200 that they had sighted 12 
ships, and everybody fell out for general 
quarters. Nothing came of this alarm and 
postwar Japanese and U. S. records in- 
dicate that there were no ships at all 
around Wake that night. 

At 0600 on 17 December Lieutenant 
Kinney reported proudly that his engi- 
neering crew had patched up two more 
airplanes. This still left the atoll with a 
four-plane air force, but fliers and other 
aviation personnel could hardly have been 
more amazed if two new fighter squadrons 
had just arrived. Major Putnam called 
the work of Kinney, Hamilton, and Hes- 
son "magical." " 

. . . With almost no tools and a complete lack 
of normal equipment, they performed all types 
of repair and replacement work. They changed 
engines and propellers from one airplane to 
another, and even completely built up new 

" Putnam Rept, 15. 



engines and propellers from scrap parts sal- 
vaged from wrecks. ... all this in spite of the 
fact that they were working with new types 
[of aircraft] with which they had no previous 
experience and were without instruction man- 
uals of any kind. . . . Their performance was 
the outstanding event of the whole campaign. ^^ 

"Engines have been traded from plane 
to plane, have been junked, stripped, re- 
built, and all but created," another report 
said of Kinney's engineering work.^^ 

At 1317 that afternoon 27 Japanese 
bombers from Roi came out of the south- 
west at 19,000 feet. Their bombs ignited 
a diesel oil tank on Wilkes and destroyed 
the defense battalion messhall as well as 
much tentage and quartermaster gear at 
Camp One. A bomb explosion also dam- 
aged one of the evaporator units upon 
which Wake depended for its water sup- 
ply. The 3-inch guns brought down one 
of these planes. 

Later that day one of the Kinney- 
patched fighter planes washed out during 
take-off, and it bad to be sent back to the 
boneyard. Then at 1750 came the heavi- 
est raid the Kawanishi flying boats ever 
put into the air against Wake." Eight of 
these planes bombed and strafed the atoll 
but inflicted little damage. 

As if the Wake defenders did not already 
have their hands full, construction author- 
ities in Pearl Harbor wanted to know how 
things were going with the lagoon dredg- 
ing. They also asked for a specific date on 
which the atoll would have certain other 

" CO VMF-211 Kept to CO MAG-21, 20Dec41, 

" As an example of how memories can grow 
dim, not a single defender remembered to men- 
tion this raid in accounts prepared after the 
war ended. Yet there is no doubt that the raid 
occurred, because the garrison reported it to 
Pearl Harbor that same afternoon. Wake File. 

improvements completed. The island com- 
mander prefaced his preliminary reply to 
this query with an account of the latest air 
raid, and followed this with a damage re- 
port which summarized his battle losses 
since the beginning of the war. He 
pointed out that half of his trucks and en- 
gineering equipment had been destroyed, 
that most of his diesel fuel and dynamite 
were gone, and that his garage, blacksmith 
shop, machine shop, and building supplies 
warehouse either had been blasted or 
burned to the ground. 

In a supplementary report sent later. 
Commander Cunningham told the Pearl 
Harbor authorities that everybody on 
Wake had been busy defending the atoll 
and keeping themselves alive. They could 
not do construction work at night, he 
pointed out, and if they used too much 
heavy equipment during the day they 
could not hear the bombers approaching. 
Besides, he reiterated, much of his equip- 
ment had been destroyed by the bombing 
raids, and most of his repair facilities had 
met the same fate. On top of all this, he 
added, civilan morale was bad. Cunning- 
ham said he could not promise a comple- 
tion date on anything unless the Japanese 
let up the pressure." The originator of 
this Pearl Harbor query might have found 
a pointed hint in this reference to a let-up 
of pressure. But at any rate Cunningham 
never again was asked how his construc- 
tion work was coming along. 

The 18th of December was quiet." One 
enemy plane was sighted in the vicinity of 


" Likewise typical of the day-to-day confusion 
which exists in the Wake records and recollec- 
tions is the fact that contemporary records — 
the Wake dispatches and Maj Bayler's official 
narrative report prepared in December 1941 — 



Wake, however, and the defenders con- 
sidered its activity ominous. It was al- 
most directly overhead at about 25,000 feet 
when first sighted. Well beyond antiair- 
craft or fighter range, it flew northwest 
along the axis of the atoll, and then turned 
south, presumably returning to Roi. De- 
fenders believed this to be a photo-recon- 
naissance flight. 

Next morning the defenders continued 
their routine work, trying to add to their 
defensive installations before the bombers 
were due. This was a routine now fa- 
miliar to them. After being cleared from 
morning general quarters, the men went 
about their work until the midday raid 
sent them to gun positions or to cover. 
After that raid was over, the men cleaned 
up after the bombs or went ahead with 
their other duties. Then late in the after- 
noon they had to take time out to deal 
with the flying boats. At night they could 
usually sleep when they were not on sentry 
duty, or standing some other type of 
Avatch. Following this pattern, crew 
members of the various batteries had com- 
pleted their sturdy emplacements, and 
everybody had contributed to the con- 
struction of primary and alternate posi- 
tions for beach defense. They had built 
more beach positions than they could pos- 
sibly man, but many of these were to be 
manned only under certain conditions.^" 
The shortage of trained fighting men was 
so critical that a well coordinated Jap- 
anese landing attack would require them 
to be everywhere at once. 

indicate that the memories of the survivors have 
almost unanimously transposed the events of 17 
and 18 December. 

'"LtCol A. A. Poindexter reply to HistSec, 
HQMC questionnaire, 8Apr47. 

At 1050, 27 bombers from Roi came in 
from the northwest at about 18,000 feet. 
They worked over the VMF-211 area 
south of the airstrip, finished off the Ma- 
rines' messhall and tentage at Camp One, 
and struck the PanAir area. Batteries D 
and E hit four of these bombers, and ob- 
servers on the atoll saw one of them splash 
after its crew bailed out over the water. 
Bomb damage at Camp One was serious, 
but elsewhere it was slight, and there were 
no casualties. 

December 20 dawned gloomily with 
heavy rain, and ceilings were low and visi- 
bility poor all day. This wide weather 
front apparently dissuaded the Japanese 
from attempting their usual noon visit, 
but it did not stop a U. S. Navy' PBY 
which arrived that day and provided 
Wake with its first physical contact with 
the friendly outer world since the start of 
the war. This plane landed in the lagoon 
at 1530 to deliver detailed information 
about the planned relief and reinforce- 
ment of the atoll. These reports con- 
tained good news for nearly everyone. 
All civilians except high-priority workers 
were to be evacuated. A Marine fighter 
squadron (VMF-221) would fly in to re- 
inforce VMF-211, which was again down 
to a single plane. And the units from the 
4th Defense Battalion would arrive on the 
Tangier to reinforce the weakened detach- 
ment of the 1st Defense Battalion. The 
PBY fliers had a copy of the Tangier's 
loading plan,-^ and this list made the ship 
seem like some fabulous floating Christ- 
mas package that was headed for the atoll. 

That night Commander Cunningham, 
Majors Devereux and Putnam, and Lieu- 

"Capt AV. S. Cunningham. USN, Itr to Capt 
S. E. Morison, USNR, 7Feb47 ; McMorris Inter- 
view, 2; CinCPac OPlan 39-^1, 15Dec41. 



tenant Commander Greey prepared re- 
ports to send back to Pearl Harbor. 
Major Bayler, his mission long since com- 
pleted, would carry the papers back as he 
complied with his orders directing him to 
return from Wake "by first available Gov- 
ernment air transportation." Mr. Heve- 
nor, the Bureau of the Budget official who 
had missed the Philippine Clipper on 8 
December, also planned to leave on the 
PBY, but someone pointed out that he 
could not travel in a Naval aircraft with- 
out parachute and Mae West, neither of 
which was available. So Mr. Hevenor 
missed another plane. 

At 0700 next morning, 21 December, the 
PBY departed. Witliin less than two 
hours, at 0850, 29 Japanese Navy attack 
bombers, covered by 18 fighters, lashed 
down at Wake through the overcast and 
bombed and strafed all battery positions. 
These were planes from Carrier Division 2 
{Soryu and Hiryu) , called in by the Japa- 
nese to help soften Wake's unexpected 
toughness.^^ Due to the low ceiling, the 
attack was consummated before the 3-inch 
batteries could get into action, but the .50 
caliber antiaircraft machine guns en- 
gaged the enemy. The attack caused little 
damage, but its implications were ominous. 

Only three hours later, 33 of the shore- 
based Japanese bombers arrived from Roi, 
and again they concentrated on Peale Is- 
land and Camp Two. They approached 
from the east at 18,000 feet in two main 
formations, and the bombs from the sec- 
ond group plastered Battery D's position 
on Peale. This unit had fired 35 rounds 
in half a minute and had hit one bomber 
when a bomb fell squarely inside the di- 
rector emplacement of the battery. This 
explosion killed the firing battery execu- 

tive, Platoon Sergeant Johnalson E. 
Wright, and wounded the range officer 
and three other Marines. 

Now there was only one firing director 
mechanism left on the atoll, and it be- 
longed to Battery E located in the crotch 
of Wake Island. But Battery E had no 
height finder, although Battery D still had 
one of these. Thus the two 3-inch batteries 
had only enough fire control equipment 
for one battery. Because of Battery E's 
more desirable location, and because it had 
escaped damage since its move to this spot, 
Major Devereux decided to maintain it as 
his primary antiaircraft defense of the 
atoll. Thus by taking over Battery D's 
height finder, certain other fire control 
gear, one gun, and the necessary personnel. 
Battery E became a fully manned and ful- 
ly equipped four-gun battery. Two other 
Battery D guns were shifted to a new po- 
sition on Peale Island where they could 
assume beach-defense missions, and the 
fourth gun remained at the original bat- 
tery position. Dummy guns also were 
mounted there to create the impression 
that the battery was still intact. As a fur- 
ther measure of deception, Battery F on 
Wilkes, also reduced to two guns, would 
open fire by local control methods when- 
ever air raids occurred. Battery D was 
parceled out tliat night, and by next 
morning the garrison on Peale had been 
reduced to less than 100 Marines and a 
small group of civilians who had been 
trained by Marine noncommissioned of- 
ficers to man one of Battery D's guns.'^ 

'■ Capture of Wake, II, 372. 

^ Of these civilians the Peale Island com- 
mander, Capt Godbold, later wrote: "The ci- 
vilians w^ho served with this battery were of 
inestimable value . . . under the capable leader- 
ship of Sergeant Bowsher, they soon were firing 
their gun in a manner comparable to the Marine- 
manned guns. Before the surrender of the is- 



By 22 December VMF-211 again had 
two airplanes capable of flight, and Cap- 
tain Freuler and Lieutenant Davidson 
took them up for morning patrol. David- 
son had been out almost an hour and was 
covering the northern approaches to Wake 
at 12,000 feet when he spotted enemy 
planes coming in. He called Captain 
Freuler, who was then south of the atoll, 
and the Marines began independent ap- 
proaches to close with the enemy. The 
Japanese flight consisted of 33 carrier at- 
tack planes (dive bombers) escorted by six 
fighters, all from the Soryu-Hiryu carrier 
division. The fighters were at 12,000 feet 
and the dive bombers at 18,000. The fight- 
ers were of a sleek new type, the first Zeros 
to be encountered over Wake. 

Captain Freuler dived his patched-up 
F4F-3 into a division of six fighters, 
downing one and scattering the others. 
Coming around quickly in a difficult op- 
posite approach, Freuler attacked another 
of the Zeros and saw it explode only 50 
feet below. This explosion temporarily 
engulfed the Grumman in a cloud of 
flames and flying fragments. The Marine 
plane was badly scorched, its manifold- 
pressure dropped, and the controls reacted 
sluggishly. As the captain turned to look 
for the atoll, he saw Lieutenant Davidson 
attacking the dive bombers. The lieuten- 
ant was diving at a retreating bomber, but 
a Zero was behind him closing on the Ma- 
rine Grumman. This enemy fighter prob- 
ably downed Davidson, because the lieu- 
tejiant did not I'eturn to the atoll. 

Meanwhile a Zero got on Freuler's tail 
while he took in Davidson's plight, and 

land, some of these men were slated to be 
evacuated to Honolulu ; however, the entire gun 
crew offered to stay on the island and serve with 
the battery." LtCol B. D. Godbold Rept, 14. 

fire from the Japanese plane wounded the 
Marine pilot in the back and shoulder. 
Freuler pushed his plane over into a steep 
dive, managed to shake off his pursuer, and 
dragged the shattered, scorched F4F into 
the field for a crash landing. In the words 
of Lieutenant Kinney, whose shoestring 
maintenance had kept VMF-211 flying for 
fifteen days: "This left us with no air- 
planes." In spite of the Marine squad- 
ron's last blaze of heroism, the enemy dive 
bombers came on in to strike at all battery 
positions. But the atoll pilots were not 
much impressed by the work of the Jap- 
anese naval aviators. "We who have been 
used to seeing only the propeller hub are a 
bit taken aback by their shallow dives and 
their inaccuracies," Lieutenant Barninger 
said. The Japanese bombs did not cause 
much damage, and there were no casual- 
ties on the ground. 

But now that carrier air was being 
brought to bear against them, the Wake 
defenders concluded that it would not be 
long before the Japanese came back with 
a bigger task force and a better amphib- 
ious plan. Ground defense preparations 
intensified that afternoon. VMF-211 's 
effectives — less than 20 officers and men — 
were added to the defense battalion as 
infantry, Peale Island completed its beach 
defense emplacements, and Captain Piatt 
drew up final detailed orders for his de- 
fense of Wilkes. Piatt ordered Marine 
Gunner McKinstry, who commanded Bat- 
tery F, to fire on enemy landing boats as 
long as his guns could depress sufficiently, 
and then to fall back to designated posi- 
tions from which his men would fight as 
riflemen. There these men from the 3-inch 
battery would be joined by the personnel 
from Battery L. After that it would be 
an infantry fight. "All that can be done 



is being done,'' noted one of the Wake 
officers, "but there is so little to do with." ^* 

11-21 DECEMBER '' 

Wake defenders were correct in assum- 
ing that the Japanese soon would be back 
with a stronger effort than the one which 
had failed for Rear Admiral Kajioka. The 
admiral began to mull a few plans for 
his next attack while he withdrew toward 
Kwajalein on 11 December, and he had 
his staif in conference on 13 December 
while his battered fleet still was anchoring 
in that Marshall atoll. Rear Admiral 
Kuniori Marushige, who had commanded 
Cruiser Division 18 (including the light 
cruisers Tatsuta and Tenryu as well as 
the flagship Yuhari) , analyzed the causes 
of failure as follows : The landing attempt 
had failed, he said, because of the vigorous 
seacoast artillery defense, fighter oppo- 
sition, adverse weather, and because of in- 
sufficient Japanese forces and means. 

But Admiral Kajioka was more inter- 
ested in the success of the next operation, 
and so was Fourth Fleet Commander 
Inouye at Truk. While the ships remain- 
ing in Kajioka's task force were being 
patched up at Kwajalein, Admiral Inouye 
sent destroyers Asanagi and Yunagi over 
to replace the destroyers lost in the Wake 
action. He also added the Ohoro, a much 
more powerful and newer ship of de- 
stroyer-leader characteristic which was 
armed with six 5-inch guns.^* The mine 
layer Tsugaru came over from Saipan 

" LtCol C. A. Barninger Rept. 

" Unless otherwise noted the material in 
this section is derived from Capture of Wake; 
Enemy Notes. 

''ONI 222-J, "A Statistical Summary of the 
Japanese Navy," 20Jul44, 56. 

with the Maizaru Special Landing Force 
Number Two/ and the transport Tenyo 
Maru and the float-plane tender Kiyo- 
kawa also joined the force. Troop re- 
hearsals began on 15 December, but Ad- 
miral Inouye still was not convinced that 
his force was large enough, and he asked 
the Commander in Chief of the Combined 
Fleet to send him more ships. 

Inouye's superior officer, now appar- 
ently convinced that Wake would be hard 
to crack, sent to the Fourth Fleet admiral 
the fleet carriers Soryu and Hiryu of Car- 
rier Division 2, heavy cruisers Aoba, Fu- 
nitaka, Kako, and Kinugasa of Cruiser 
Division 6, heavy cruisers Tone and Chik- 
uma of Cruiser Division 8, and a task 
force screen of six destroyers." The com- 
mander of the Combined Fleet assigned 
over-all command of this Wake task force 
to Rear Admiral Koki Abe, commander of 
Cruiser Division 8. Rear Admiral Kaji- 
oka retained his command of the amphib- 
ious force. 

Plans for the second attack against the 
American atoll called for more softening 
up than Wake had received previous to 
Kajioka's first attempt to land troops 
there. On 21 December, two days prior 
to the proposed landing, the aircraft of 
Carrier Division 2 would work over the 
atoll's defenses to destroy first the U. S. 
air capability and then the shore batteries 
and the antiaircraft weapons. Then the 
amphibious force would move up for the 
landing, and in order that the atoll might 
be surprised ^® there would be no pre- 
liminary naval bombardment on 23 De- 

To make sure that troops got ashore, the 
two destroyer-transports {Patrol Craft 

Ibid., 47, 49. 
'Capture of Wake, II, 372. 



32 and 33) would run aground on the 
south shore of the atoll near the airstrip, 
and the approximately 1,000 men of the 
special naval landing force ^^ would then 
be carried to the beach in four to six land- 
ing barges. Two of these would land on 
Wilkes Island, two on Wake Island be- 
tween the airstrip and Camp One, and the 
other two probably provided for would 
put their troops ashore just west of Pea- 
cock Point.^" If these special landing 
force troops ran into serious trouble on the 
atoll, the naval force would send in 500 
men organized from ships' landing forces. 
And if this combined force failed to sub- 
due the atoll defenders, more help would 
be sent by means of an ultimate and des- 
perate expedient. The destroyers of the 
task force would be beached, and their 
crews would swarm ashore. Admiral 
Inouye was determined that this second 
attack should not fail. 

The possibility of U. S. naval surface 
intervention was taken into consideration. 
This possibility had been dismissed during 
planning for the attack of 11 December 
because the Japanese reasoned that the 
shock of Pearl Harbor would immobilize 
American surface operations for some 
time. But now the Japanese assumed that 
U. S. surface opposition was probable. To 
guard against such threat, the four heavy 
cruisers of Cruiser Division 6 would act 
as a covering force east of Wake. If a 
major surface action developed, Rear Ad- 
miral Abe would enter the fight with 
Cruiser Division 8 and conduct the battle. 
As on the first attempt, submarines would 
precede the invasion force to reconnoiter 

the island and to look out for U. S. sur- 
face forces. 

With these final plans issued, the in- 
vasion force well rehearsed, and carriers 
Soryu and Hiryu on their way down from 
north of Midway, the operation against 
Wake was ready to go. At 0900 on 21 
December Admiral Kajioka cleared Eoi 
with the ships of his amphibious force and 
headed back up toward the American-held 

15-23 DECEMBER^^ 

Now the U. S. commanders taking help 
to Wake were in a race with Admiral 
Kajioka, even if they did not know it. 
Admiral Fletcher's Task Force 14 sortied 
from Pearl Harbor in two task groups on 
15 and 16 December, ^^ rendezvoused 
southwest of Oahu during the afternoon 
of this second day, and sailed westward to- 
ward Wake. Fletcher's force was to ar- 
rive at the atoll on 23 December (east 
longitude time). There the pilots of 
Major Verne J. McCaul's VMF-221 
would fly in from the carrier Saratoga 
while the Tangier anchored off Wilkes 
channel to unload supplies, equipment, 
and the Marines from the 4th Defense 
Battalion.^^ After taking wounded men 

"^ Wake Attack, 2. 

'° HistSec, HQMC interview with Col J. P. S. 
Devereux, 12Feb47, 8, hereinafter cited as 
Devereux Interview. 

"' Principal sources bearing on the Relief Ex- 
pedition are: CinCPac OPlan 39-41, 15Dec47; 
McMorris Interview; Saratoga log; LtCol R. D. 
Heinl, Jr., "We're Headed for AVake," MC Ga- 
zette, June 1946. 

'^ Dates in this section are either west or east 
longitude as applicable to the location of events 
identified by dates. Where confusion is possible 
the type of date will be indicated. 

^ Troops and equipment would be transported 
to the atoll on lighters, and if the Tangier were 
seriously damaged by enemy action she would be 
run aground so the cargo would not be lost. 



and certain civilians on board, these ships 
then would return to Pearl Harbor. 

But the advance was aggravatingly slow. 
The old fleet oiler Neches^ in the train with 
the Tangier, could not manage more than 
12 knots, and the fleet's zig-zag evasive 
tactics further slowed the rate of advance. 
To the Marines and seamen in this first 
westward sally of the war, the waters be- 
yond Oahu seemed very lonely and omi- 
nous, and there was no contact, either 
friendly or enemy, to vary the tense 
monotony of the run. Each day on the 
Tangier began with general quarters and 
then lapsed into normal shipboard routine. 
Marines received such training and in- 
struction as could be fitted into this sched- 
ule, and part of this educational program 
included lectures by the few radar techni- 
cians on board. 

The few available maps and charts of 
Wake received intense study. In antici- 
pation that Wake's 3-inch guns might have 
to deliver direct fire on ships or ground 
targets, improvised sights were designed 
and constructed in the ship's machine 
shops. The officer commanding the ma- 
chine-gun detachment contrived with the 
ship's force to construct special slings with 
which his .50 caliber antiaircraft machine 
guns could be hoisted from ship to barges 
while remaining ready to ward off possible 
enemy attacks during unloading. ' The 5- 
inch seacoast men stayed in practice by 
standing their share of watches on the 
after 5-inch gun of the Tangier. All Ma- 
rine antiarcraft machine guns were set up 
and manned on the superstructure. 

On 18 December CinCPac ordered U. S. 
submarines which were patrolling in the 
vicinity of Wake to move south out of the 
area. These boats of Task Force 7 were 

to patrol around Rongelap in the Mar- 
shalls until the relief expedition reached 
Wake. CinCPac wanted to avoid any pos- 
sibility of one U. S. force confusing 
another for the enemy.^* Three days later, 
on the 21stj intelligence information which 
had been arriving at Pearl Harbor indi- 
cated that a large force of shore-based 
Japanese planes was building up in the 
Marshalls, and that enemy surface forces 
might be east of Wake where they could 
detect the approach of Fletcher's Task 
Force 14. Other reports indicated the 
presence of Japanese carriers, including 
possibly the Soryu, northwest of the atoll. 
Fletcher's mission, now about 650 miles 
east of Wake, appeared to be growing 
more hazardous with each hour. CinCPac 
ordered the carrier Lexington and other 
ships of Task Force 11 over from the 
southeast to give Fletcher closer support.^^ 
By 0800 on 22 December, Task Force 14 
was within 515 miles of Wake, and Ad- 
miral Fletcher in the cruiser Astoria kept 
up on the news about his race by monitor- 
ing the CinCPac radio nets. Ominous re- 
ports of Japanese surface operations 
around the atoll continued to filter in at 
Pearl Harbor, but conditions at Wake were 
unchanged. Fletcher decided to refuel. 
Although his destroyers still had a reason- 
able supply of oil, it might not be enough 
if they had to fight. But this very act of 
fueling, which took most of the day, kept 
them out of the fight. By the time the 
U. S. ships moved on toward Wake, Ad- 
miral Kajioka was only about 50 miles 
from the atoll with his amphibious force. 
Fletcher had lost the race. 

'* Paraphrased file of CinCPac dispatches con- 
cerning Wal?e relief, December 1941. 

The Fall of Wake ' 


At Wake, 23 December began with inter- 
mittent rain squalls, and shortly after 0100 
the defenders saw a succession of vivid, ir- 
regular flashes beyond the horizon north 
of Peale Island. Men on the atoll could 
hear nothing above the rain and the boom 
of the surf, but it was obvious that the 
flashes were not signals or searchlights. 
They were too brilliant and irregular for 
that. Old fleet-duty hands were reminded 
of night battle practice at sea. Was there 
a naval battle, or were the Japanese com- 
ing back ? The defenders could only 

By this time the Marines were used to 
seeing lights, even though these were un- 
usual. But at 0145 came a more urgent 
alarm. The word over the "J"'-line an- 
nounced that the Japanese were landing 
at Toki Point on Peale. Major Devereux 
alerted all units and then telephoned 
Lieutenant Kessler at Toki Point for ad- 
ditional information. The Battery B 
commander told Devereux that he could 
see lights in the distance but that there 
was no landing in progress. The beach 
positions had been manned, Kessler added, 
because boats were "believed" to be some- 

' Unless otherwise noted the material in Chap 
4 is derived from Dcrcrciix Rcpt ; Putnam Repi ; 
(Officer's name) Repts, especially LtCol W. McC. 
Piatt Rept, IstLt J. A. McAlister Rept, MG C. B. 
McKinstry Rcpt; Devereux Intervicic : replies 
to HistSec questionaire by Col G. H. Potter, 
LtCol W. McC. Piatt. LtCol A. A. Poindexter : 
Capture of Wake; Wake Attack; Hawaii- 
Malaya Na^-Ops; Devereux Story. 

where offshore. By this time all units 
had sent their men to general quarters, 
and at Camp One, Second Lieutenant 
Poindexter loaded his scanty mobile re- 
serve unit of eight Marines ^ and four .30 
caliber machine guns into their truck, re- 
ported his actions to the command post, 
and moved out toward Peale Island. But 
the word from Kessler had convinced 
Devereux that if the enemy were landing, 
they were not doing it on Peale Island. 
He put a damper on the general alarm, 
and ordered that Poindexter be inter- 
cepted when his truck passed the com- 
mand post. He held the mobile reserve 
there to await developments. 


Developments were not long in coming. 
Admiral Kajioka's amphibious force had 
at last sighted the atoll's faint outline, 
and the ships were reducing speed. Mo- 
ments later, in the words of a Japanese 
"combat correspondent" who was moved 
to poetry by this amphibious venture, 
"The honorable, first order of 'CHARGE' 
was given, and the daring officers and men, 
with white sashes, bravely went down to 
the surface of the sea." ^ 

" This unit, along with a few Marines from 
supply and administration sections and 1.5 Navy 
enlisted men commanded by BM IstCl James E. 
Barnes, was also responsible for the defense of 
Camp One. 

^ This latter-day Masefield was Kayoshi Ibushi 
of the .Japanese Naval Information Section. 




Approximately 800 of the SNLF troops 
were distributed between the two de- 
stroyer-transports.* The other 200 pre- 
sumably were embarked on board one or 
more of the transports or the float-plane 
tender Kiyokawa^ and the 500 sailors of 
the provisional reserve force apparently 
were to remain at their normal duties un- 
less called to reinforce the landing effort.^ 
The Maizwu Second Special Naval Land- 
ing Force^ now brought to full strength 
by reinforcements from Saipan, was es- 
sentially a Japanese version of the bat- 
talion landing team (BLT). Its three 
rifle companies had numerical designa- 
tions but were more commonly identified 
by the names of their commanders. Thus 
the 1st Company^ commanded by Lieu- 
tenant Kinichi Uchida, was often called 
the Uchida unit. Similarly, the 2d and 
3d Companies wore styled respectively the 
Takano unit and the Itaya unit. 

The Uchida and Itaya coinpanies would 
assault Wake Island while 100 "picked 
men" of the Takano unit seized Wilkes."' 

"This reporter," as he later said of himself, 
". . . was able to have the honor of taking the 
first step upon the island as a man of letters . . . 
the capture of AVake Island . . . was so heroic 
that even the gods wept." Hawaii-Malaya Nav- 
Ops, 32. 

* One account says 500 men of this force were 
on the two destroyer-transports. At this time 
all SNLF units attached to the Fourth Fleet 
were concentrated for the seizure of Wake in 
much the same manner that U. S. Fleet Marine 
Force units would have been employed. 

' The exact number of Japanese troops who 
fought on- Wake has not been determined. Adm 
Morison cites Marine estimates that "at least 
1,200 troops landed early on the 23rd," and that 
others came ashore after the surrender. Rising 
Sun in the Pacific, 24.5. 

" Itaya's name and that of his company ap- 
pears in one Japanese source as "Itatani." 
Hawaii-Malaya NavOps, 28. 

The balance of the Takano company pre- 
sumably would back up the other two 
companies on Wake Island. At about the 
time the premature landing alarm was 
sounded on the atoll, the amphibious force 
was putting landing craft over the side. 
The weather was giving them trouble, but 
at about 0200 the SNLF troops clambered 
down into these craft. "The hardships en- 
countered in lowering the landing barges 
were too severe even to imagine," reported 
correspondent Ibushi. "Now we, the Na- 
val Landing Force, on the barges which 
we were in, must charge into enemy ter- 
ritory and carry out the final step of se- 
curing a landing point after touching the 
shore." ^ 

As the landing craft pitched through 
the breakers, the destroyer-transports 
turned to make their final runs onto the 
reef south of the airstrip. These vessels. 
Patrol Boats 32 and 33, mounted the reef 
in a smother of breakers and foam, and 
went aground near the west end of the air- 
strip. Two of the landing barges scraped 
bottom as they approached the reef near 
Camp One, and still there was no sign that 
the atoll defenders were awake. But sud- 
denly tracers pencilled from the beach at 
Wilkes Island and .50 caliber slugs splat- 
tered through the gunwales of one barge. 
Then a searchlight from Wilkes flared on 
to silhouette the picked men of the Takano 
unit landing on that island. It was then 
0245, and the battle for Wake was on. 
(See Map 5, Map Section) 


Since 0215 Marines had been confident 
that a landing against them was in prog- 




ress. Lights could be seen offshore iiortli 
of Peale Island and all along the south 
coasts of Wake and Wilkes Island. At 
about 0230 Marines on Peacock Point 
thought they could see the outlines of two 
barges heading along the coast toward the 
airfield, but these evidently were the pa- 
trol craft heading in toward the reef. By 
now^ Major Devereux, Major Potter, his 
executive officer, a radioman, and a 
switchboard operator in the defense de- 
tachment command post were swamped by 
reports of sounds, lights, and shapes. As 
he collected this information and relayed 
reports to Commander Cunningham, 
Devereux saw that the greatest threat was 
developing along the south coast of the 
atoll, and he dispatched Lieutenant Poin- 
dexter's eight-man mobile reserve to de- 
fensive positions between Camp One and 
the airstrip. 

Poindexter's men had not left the truck, 
and the lieutenant had them transported 
down the island and into position within 
15 minutes. The area into which they 
moved was just west of the road junction 
near the west end of the airstrip. There 
this small force commanded the south 
shore road as well as the critical beach 
section south of the field. The lieutenant 
reported that this area was being bom- 
barded when he reached his defensive po- 
sitions, but there were no signs of a 

But at 0235 ^ defenders on Wilkes re- 
ported that they could hear barge engines 

" The Japanese list 023.5 as the official time of 
their landing. Statements of Marine officers do 
not agree. Some say the landings came at about 
0130, while others place the time almost an hour 
later. In his official report, Devereux gives 
the time of landing on Wilkes at 021.5, but his 
published narrative says 0120. Yet a subsequent 
study convinced him that the time of 0235 was 

above the surf, and Marine Gunner Mc- 
Kinstry opened fire with a .50 caliber ma- 
chine gun at a dark shape near the beach 
below. Ten minutes later Captain Piatt 
requested permission to illuminate the 
beach with his 60-inch searchlight, and 
the landing was discovered. Two barges 
could be seen on the beaches at Wilkes, 
the lights also revealed the patrol craft 
aground off Wake. 

Neither of the 5-inch batteries which 
commanded the south approaches to 
Wake ^ could bear against the landings. 
Terrain masks likewise prevented them 
from firing at Patrol Craft 32 and 33 on 
the reef.^" The only weapon larger than 
a machine gun that could engage these de- 
stroyer-transports, already beginning to 
spew out their human cargo, was the 3-inch 
gun emplaced on the rise between the beach 
road and VMF-211's hard-stand parking 
area. But this gun was not manned. 
Realizing the importance of this weapon, 
Second Lieutenant Robert M. Hanna, in 
command of the antiaircraft machine guns 
about the field, gathered a scratch crew 
consisting of one Marine, Corporal Ralph 
J. Holewinski, and three civilians " and 

closer to the fact. Excepting the variances con- 
cerning events during these dark early hours of 
the battle, Marine accounts agree generally as to 
events after daylight. 

' Btry A (Peacock Point) and Btry L (Wilkes). 

" One of the advantages of the Navy 5"/51 for 
seacoast defense missions was that it had 360° 
train, as contrasted to the limited traverse of 
the 155mm field guns used by the Army for this 
role. In this instance, however, terrain masks, 
slight though they were, prevented either A or 
L from bearing. 

" IstLt R. M. Hanna Kept to CO, 1st DefBnDet, 
110ct45, 2. Of these three civilians, two (Paul 
Gay and Bob Bryan) were subsequently killed 
in action, and the third, Eric Lehtola, was 
wounded. Hanna states that they fought with 
"exceptional gallantry." 



JAPANESE PATROL CRAFT lost in the assault on Wake Island are silent witnesses to an American 
carrier raid during the last days of the war. ( USN 495560 ) 

JAPANESE NAVAL TROOPS who took Wake Atoll are shown in a contemporary propaganda painting 
taking their prisoners toward a collecting point. (SC 301066) 



raced to this gun. Major Devereux also 
realized the critical importance of holding 
this area, and he ordered Major Putnam 
and the 20 men of VMF-211 to form an 
infantry support between the 3-inch gun 
and the enemy landing. 

All defense units on Wake Island were 
disposed to meet the enemy. Hanna and 
the VMF-211 "infantrymen" held the 
left flank south of the airfield parking area. 
To the west, and squarely in the path of 
the enemy's initial rush toward the west 
end of the field, were Second Lieutenant 
Kliewer and three aviation Marines. They 
guarded one of the generators which was 
wired to detonate the mines buried in the 
airstrip. At the road junction farther 
west Poindexter's mobile reserve was 
already firing its four machine guns east- 
ward along the beach at Patrol Craft 32 
where the enemy troops had revealed them- 
selves by injudicious use of pyrotechnic 
signals. At Camp One four .30 caliber 
machine guns were manned for beach de- 
fense by Battery I's gun shed crew and the 
Naval Air Station sailors who had been 
serving as lookouts on the water tank OP. 
Behind this general line, two .50 caliber 
machine-gun sections (each of two guns) 
guarded the airstrip. One section held 
the west end of the strip near Lieutenant 
Kliewer's generator, and the other section 
was located on the east end of the strip.^^ 
These two sections could command the 
length of the field, and could partially in- 
terdict movement across the field. Other 

"These two sections had been sited to provide 
antiaircraft Are as well as final protective fire 
along the airstrip. This conformed with defense 
battalion practice, but the light of hindsight 
prompted Devereux to wish that he had moved 
these sections closer to the beach where they 
could have been tied in with the general line. 

machine guns were in the Peacock Point 
area. At the battery positions gun crews 
stood by their weapons and manned such 
local perimeter defenses as their meager 
strength permitted. 

Lieutenant Hanna and his jury-rigged 
crew quickly got the 3-inch gun into ac- 
tion. They laid the weapon by estimate 
and "Kentucky windage"," and fired their 
first round at Patrol Craft 33 which was 
less than 500 yards away. The shell hit 
the bridge of the destroyer-transport, and 
wounded the captain, the navigator, and 
five seamen. Two other sailors were killed. 
While men of the Uchida and Itaya wnits 
swarmed off the ship, Hanna and his crew 
fired 14 more rounds into the superstruc- 
ture and hull of the craft. Finally it burst 
into flames, illuminating the landing area. 
"The scene was too beautiful to be a battle- 
field," reported a Japanese observer on 
board the cruiser Yuha?^.^* 

Flames from tliis ship lighted Patrol 
Craft 32 farther west along the beach, and 
Hanna shifted his fire to this vessel. 
Three-inch shells hulled this transport-de- 
stroyer, and crews from both these ships 
joined the SNLF troops landing on the 
island. This added possibly 100 extra 
men to the battle ashore, and Hanna's gun 
already was seriously threatened by the 
Uchida unit which had made the beach 
assault. Major Putnam's aviators fought 
oflF these early attempts to silence the 3-inch 
gun, but the Japanese continued to attack. 
Alternating between creeping infiltration 
tactics and screaming rushes, the I'chida 
troops drove the Marines back on each side 

'' These 3-inch Army antiaircraft guns were 
equipped only for indirect fire at aerial targets, 
and they had no sights or other fire control equip- 
ment to facilitate direct sighting by local control. 

" Haicaii-Malaya NavOps, 33. 



of the gun until the defenders were in a 
position which Devereux later described 
as "a box-shaped thing.'' Here they con- 
tinued to hold. 

But farther west Japanese troops had 
reached the south shore road between the 
mine field generator and Lieutenant Poin- 
dexter's small mobile reserve, and by the 
light from the burning patrol craft Poin- 
dexter could see these enemy cross the road 
and disappear into the brush beyond. 
The lieutenant was directing machine- 
gun fire into this brush area when he 
heard other firing from the direction of 
Camp One. He left Gunnery Sergeant 
T. Q. Wade in charge of the reserve force 
and headed toward the camp. There he 
found that two large landing craft had 
grounded on the reef some 30 yards off- 
shore southeast of the camp." Four ma- 
chine guns from Camp One fired at the 
barges, but the rounds ricocheted off. This 
fire evidently discouraged the craft from 
attempting a landing at this point, how- 
ever, because they backed off the reef and 
nosed about as if seeking a better site. 

While these two boats floundered about 
in the surf, Lieutenant Poindexter formed 
two teams of grenadiers to move down to 
the water's edge and lob hand grenades 
into the barges. One team consisted of 
himself and Boatswain's Mate First Class 
James E. Barnes, while the other consisted 
of Mess Sergeant Gerald Carr, and a civil- 
ian, K. R. Rutledge, who had served as an 
Army officer in France during World 
War I. The machine guns suspended fire, 
and the grenadiers attacked. By this 
time the Japanese had landed a short dis- 
tance farther east, and Boatswain's Mate 
Barnes managed to throw at least one 
grenade inside a barge just as the enemy 


debarkation commenced. The explosion 
inflicted heavy casualties, but some 75 to 
100 enemy splashed ashore and entered the 
underbrush east of Camp One. This heavy 
growth north of the road soon became a 
sort of no man's land into which the Japa- 
nese continued to infiltrate and expand 
their beachhead. 

All this Poindexter managed to report 
back to Major Devereux in a final message 
from Camp One before wire communica- 
tion was lost. But shortly after this a 
panicky civilian who had managed to pick 
his way through the brush from Camp One 
to Devereux's command post brought in 
reports — totally untrue — that Camp One 
was being overrun and that he had seen 
Japanese troops bayoneting the machine 
gunners of the mobile reserve. 

The loss of communications was not 
localized at Camp One. Devereux's com- 
mand post had lost contact with Lieuten- 
ant Hanna, the VMF-211 infantrymen, 
the CP of the .50 caliber machine-gun bat- 
tery near the airstrip, and Battery A at 
Peacock Point. The tactical line to 
Wilkes Island also went out at this time, 
but the "J"-line, which lay north of the 
airfield, still linked the defense battalion 
CP with that of Captain Piatt on Wilkes. 
Nobody knows exactly what caused this 
communications failure, but the nature of 
the trouble suggests that it might have 
been caused by a single break. The loca- 
tion of this major break, if there was one, 
must have been near the battalion com- 
mand post where lines were close together. 
But all Wake survivors hold the opinion 
that the Japanese cut the lines; and they 
point out that the Wilkes "J"-line did not 
go out until some time after this failure 
of the line south of the field. Thus de- 
fenders believe that the lines were being 



cut as the enemy attack progressed in- 
land. Devereux tried to contact his Wake 
Island positions by radio, but this inter- 
island net never had been reliable, and the 
sets characteristically failed to function 
that morning. There wer6 no communi- 
cations personnel in the command post to 
trouble-shoot lines, " and for the remain- 
der of the battle Devereux had no com- 
munications with his defensive line along 
Wake's south coast. 

It was now obvious to Devereux at his 
"blacked-out" CP that fights were in prog- 
ress all along the west leg of Wake Is- 
land, and that he must sacrifice a defensive 
unit from some other area to reinforce his 
effort in the critical zone. Lieutenant 
Lewis' Battery E in the crotch of Wake 
Island could not be disbanded. It was the 
only completely equipped and up-to- 
strength ^' antiaircraft battery on the 
atoll. Battery B's 5-inch guns on Peale 
Island also should remain manned for pos- 
sible missions against enemy ships. But 
Captain Godbold's Battery D might be 
used as infantry. This unit had two 3- 
inch guns, but no fire control equipment; 
and Peale Island did not appear to be 
threatened. Two officers and some 40 men 
from this battery became the atoll reserve, 
and at 0300 Major Devereux ordered God- 
bold to send one gun section (about nine 
men) from this reserve force to the aid 
of Hanna's untrained crew. Corporal 
Leon Graves brought these men around 
from Peale Island in a truck driven by 

'"MSgt R. M. June reply to HistSec, HQMC, 
questionnaire, llMar47, 2. 

" "Up-to-strength" was a relative term on 
Wake. The 1941 tables of organization allowed 
such batteries two oflScers and 75 enlisted men. 
At this time Battery E had two oflScers and 
about 50 men. 

one of the civilians, and Major Devereux 
directed them toward Hanna's position. 

By this time the Japanese had made at 
least two penetrations through the Marine 
"line" along the south edge of Wake Is- 
land, and it is possible that the enemy was 
also landing inside the lagoon in rubber 
boats. Several defenders speak of seeing 
red flares rising from within the lagoon, 
and after the surrender Marine working 
parties found rubber boats on these in- 
terior beaches. If these landings were 
taking place, it is probable that they oc- 
curred on the north beaches of Wake Is- 
land's west leg.^* From such sites the men 
landing in rubber boats could join up with 
those landing on the south beaches. 

Captain Godbold on Peale Island was 
one of the defenders who saw these red 
flares inside the lagoon, and he had Bat- 
tery B at Toki Point send a two-man 
patrol down the interior coast of that 
island to investigate. Godbold then sent 
a three-man patrol from his own battery 
down the outer coast of Peale. These two 
patrols met at the southeast end of the 
island without encountering any enemy. 
The captain then established a three-man 
outpost to cover the bridge between Peale 
and Wake Islands. 

^ Although Japanese sources do not mention 
such interior landings the evidence to support 
them is generally convincing. The rubber boats 
did not enter the lagoon through the channel be- 
tween Wilkes and Wake Island, because this 
narrow channel was covered throughout the bat- 
tle. Devereux surmises that the boats entered 
the lagoon at the open end of the atoll between 
Kuku and Toki Points. Such landings would 
explain, without necessarily ruling out infiltra- 
tion, the early presence of individual Japanese 
at various points along the lagoon shore. One 
Japanese source does mention that red rocket 
flares were to be used as a signal that "We have 
succeeded in landing." Hawaii-Malaya NavOps, 



But interior landings or no, the Wake 
defenders had their hands fulL Japanese 
cruisers began to bombard the atoll's main 
island at about 0330. The landings con- 
tinued in spite of the fact that Battei-y E 
now fired air bursts over the beaches, and 
enemy infantry continued to press closer 
to Hanna's 3-inch gun south of the air- 
strip. The VMF-211 troops still held, 
but their partial perimeter was being com- 
pressed tighter and tighter around the 
gun. This action was now little more 
than a battle for preservation of the 
weapon and the Marines involved. Major 
Putnam's men could not check the Jap- 
anese penetration farther to the west, nor 
could they prevent the enemy from mov- 
ing behind them or into the island tri- 
angle above Peacock Point. And the Jap- 
anese wanted to concentrate in this tri- 
angle so they could launch an attack up 
the island's east leg. The VMF-211 
troops could only hope to cling to the 
slight hillock of their position, and stay 
there as long as possible.^** 

Meanwhile Coporal Graves and his de- 
tached gun squad from Battery D were 
trying to reach Hanna's 3-inch gun. 
Devereux had told them to detruck at the 
road junction some 600 yards below the 
end of the strip and west of Peacock 
Point. From there they were to go 
through the underbrush to the gun posi- 
tion. But the squad detrucked consid- 
erably short of this junction — prpbably 
less than 200 yards below the strip. From 
there the men struck out through the brush 
in the general direction of the Hanna- 
VMF-211 area. They were soon stopped, 

however, by enemy machine-gun and 
small-arms fire which killed one Marine 
and pinned down the others.^" After a 
time Graves withdrew his unit northward 
toward the command post where it later 
participated in defensive efforts com- 
manded by Major Potter. 

It is not clear what sort of an enemy 
force Corporal Graves encountered in the 
Peacock triangle, or how the Japanese got 
there. There are indications that a land- 
ing might have been made in that area, 
with barges coming in on the south coast 
between Battery A on Peacock Point and 
the Hanna- VMF-211 position. Dev- 
ereux said after the war that he believed 
a landing took place at this point, bvit the 
matter never has been confirmed. Some 
Jai^anese accounts, including those of 
Captain Koyama and a correspondent,^^ 
mention a landing "near the southeast tip 
of Wake" to overrun Battery A, which 
must have been remembered from the ac- 
tion of 11 December — especially by men 
in the cruiser Ynhari. But Captain Ko- 
yama also insisted that the Japanese made 
only two barge landings with a total of 
four barges. And these are accounted for 
by the landings near Camp One and at 
Wilkes Island. Discounting a third barge 
landing, this force must have been built 
up by the rubber boat landings within the 
lagoon, or by wholesale infiltration behind 
the position held by Putnam and Hanna. 

But at any rate, Devereux soon learned 
from Corporal Graves that there was an 

^'Kinney Interview. At about this time Maj 
Putnam, already wounded, told his men, "This 
is as far as we go." Six hours later, when the 
island fell, they still held. 

^° Col Devereux suggests that some of the 
machine-gun fire which swept through the Pea- 
cock triangle might have come from friendly 
weapons. He points out that Marines had .30 or 
.50 caliber machine gun sections on virtually the 
entire perimeter of the triangle. 

" Capture of Wake, II, 372. 



enemy force in the triangle. And from 
there the Japanese threatened the entire 
eastern rim of the atoll. Battery E was 
now receiving light mortar and long-range 
machine-gun fire, and Battery A likewise 
began to receive enemy mortar fire.-^ In 
the face of this, Captain Barninger armed 
his range section with two .80 caliber ma- 
chine guns and formed an infantry out- 
post line on the high ground behind his 
5-inch guns. 

The enemy fire against Battery E 
seemed to come from the thick brush on 
the other side of an inlet southwest of the 
battery position. Direct 3-inch fire into 
this area silenced one automatic weapon, 
but this did not seem to ease the pressure 
much. Lieutenant Lewis then sent a 
patrol of approximately 10 men under 
Sergeant Raymon Gragg to investigate. 
Gragg went out to the road north of the 
airstrip, and patrolled to the southwest 
along this road. Within 50 yards of the 
battery Gragg's patrol ran into heavy 
Japanese fire which forced the Marines to 
deploy. Answering the enemy fire, the 
patrol held here until the surrender. 

At about 0430 the .50 caliber machine- 
gun section at the east end of the airstrip, 
still in communication with Devereux, re- 
ported that the Japanese were attacking in 
company strength up the road from Pea- 
cock Point. Corporal Winford J. Mc- 
Anally, in charge of the six Marines and 
three civilian volunteers at this position, 
was trying to hold the Japanese south of 
the airstrip. Fire from the .50 caliber gun 
position had halted the enemy advance 

"^ Barninger's report also speaks of occasional 
fire from "a small field piece." Lt Col C. A. 
Barninger Rcpt, 6. This may have been a 70mm 
howitzer of the type organic to Japanese infantry 

along the road, but the enemy now at- 
tempted to infiltrate around the strong- 
point. McAnally contacted another ma- 
chine gun position some 400 yards to the 
south on the atoll's east shore, and these 
two sections alternated in firing at the 

This Japanese force probably was the 
Itaya unit. This reinforced company 
evidently infiltrated behind the Putnam- 
Hanna position at the 3-inch gun while the 
Uchida company remained near the beach 
to deal with that weapon which had fired 
on the patrol craft. The enemy at first 
had trouble locating McAnally's gun sec- 
tion, but before daylight they were all 
around the position. McAnally "s men con- 
tinued to hold, however, and the corporal's 
reports to Devereux gave the major his 
only link with the action south of the com- 
mand post. 

By 0500, a half hour before dawn, it 
was clear that the Japanese had a superior 
force firmly established on the atoll, and 
that the enemy was free to infiltrate almost 
at will around and between the isolated 
positions of the defenders. At this time 
Commander Cunningham sent his mes- 
sage, "Enemy on island issue in doubt." ^^ 
But actually there was little doubt, al- 
though the defenders were far from ad- 
mitting it at that point. The 500 defend- 
ers on the atoll were then outnumbered ap- 
proximately two to one by the enemy ; but 
what was worse, the Marines had their 
mission and their own atoll against them. 
"Little Wake" has a vulnerable shore line 
about 21 miles in length, and the defenders 
had insufficient men to man even a mini- 
mum of their antiaircraft and seacoast 
guns and at the same time the beach de- 
fenses. On Wake Island alone, nearly 

Wake File. 



half of the 200 defenders had to remain at 
Batteries A and E, and another 15 Ma- 
rines manned machine guns and search- 
lights at Heel Point where the island's east 
leg crooks toward Camp Two. Thus only 
about 85 men could oppose the enemy land- 
ing force, and half of these were machine- 
gun crewmen. Marines serving as rifle- 
men against the enemy on Wake Island 
numbered between 40 and 45. 

When Cunningham sent his message, 
Major Putnam still held the position 
around Hanna's gun, but the Japanese now 
had these Marines surrounded. Here the 
defenders had sustained a number of cas- 
ualties, including the death of Captain 
Elrod.^* Camp One also continued to 
hold, and Lieutenant Poindexter had re- 
joined his small mobile reserve force near 
the road junction west of the airstrip. 
There at first dawn the Marines were taken 
under heavy fire from the brush off their 
left (north) flank. Light mortar shells 
began to fall around the gun positions, and 
one of the .30 caliber weapons was put out 
of action. In danger of being outflanked 
here, Poindexter ordered a withdrawal to 
Camp One where he would consolidate for 
his final stand. The unit displaced by sec- 
tion in 150-yard bounds, and arrived at 
Camp One shortly after daybreak. There 
Poindexter organized his defenders along 
a semi-circular line facing seaward and 
to the southeast. In this line he had about 
40 riflemen and 10 machine guns. 

Lieutenant Kliewer and his three Ma- 
rines had survived the night beside the 
mine field generator, but a heavy Japanese 
attack threatened them just before dawn. 

'* The Japanese sustained at least 62 casualties 
trying to take this gun position, and one of them 
was Lt Uchida, the company commander. 

This was repulsed by close-in fighting with 
submachine guns and grenades, but the 
Japanese came back again at dawn. This 
time the enemy made a shouting bayonet 
charge against the Marines, but again 
Kliewer and his men, now aided by the .50 
caliber machine guns at the west end of 
the airstrip, managed to halt the attack. 

Enemy pressure against McAnally's ma- 
chine-gun position east of the airstrip also 
increased during the hour before dawn. 
The Marine strong point now had been 
located, and the defenders were under 
heavy attack by small-arms fire and gren- 
ades. McAnally's gunners already had 
broken up a number of enemy rushes by 
holding their fire until it would be most 
effective, but these 10 men could not ex- 
pect to hold out for long against the 
reinforced company opposing them. 

This was clear also to Devereux at the 
command post, and at 0530 he directed 
Major Potter, who until now had assisted 
in the command post, to assemble every 
headquarters, service, supply, or casual 
Marine in the command post area, includ- 
ing Corporal Graves' detached squad from 
Battery D, and to form a final defensive 
line approximately 100 yards south of the 
command post. This force of approxi- 
mately 40 men would take up positions 
astride the north-south main road. Dev- 
ereux then telephoned Captain Godbold on 
Peale and directed him to truck his entire 
Battery D, plus the few .50 caliber gun- 
ners, to the battalion command post for 
immediate employment as infantry. With 
these orders, the atoll's final reserve, total- 
ling approximately 30 officers and men, 
was committed. 

By 0600 McAnally's position was nearly 
surrounded and under continual infan- 



try ^* attack. Unless he was to lose these 
personnel, Major Devereux had no alter- 
native but to pull them back. This he did 
shortly after 0600, when McAnally was 
ordered to withdraw northward and join 
Major Potter's line. 

After Captain Godbold's reserve force 
left Peale Island, First Lieutenant Kessler 
became commander there since his Bat- 
tery B was all that remained on the island. 
In the light of dawn Kessler could see on 
Wilkes a line of Japanese flags across the 
center of the island, and a large enemy 
flag waving from the approximate position 
of Marine Gunner McKinstry's provi- 
sional Battery F. This he reported to 
Major Devereux, who could only conclude 
that Wilkes, which had been silent since 
about 0300, had shared the fate which now 
appeared imminent for Wake Island. 

Above the brush and slight rise of 
ground which topped the west leg of Wake 
Island, Kessler could also see the super- 
structure of Patrol Craft 32. Observing 
that the ship appeared intact, Kessler at 
0600 requested Major Devereux 's permis- 
sion to fire on it. Although the line of 
fire and intervening partial mask ^ made 
this hazardous, the request was approved, 
and on the first salvo Battery B shot away 
the ship's mainmast. As a result of sub- 
sequent adjustment, the ship was hit about 
the superstructure and upper hull. It 
finally caught fire. 

^ Among the Japanese killed before his position 
at about this time were two flame-thrower oper- 
ators. Although use of flame is not recorded, 
this was perhaps the earliest tactical employ- 
ment of this \vea])on in the Pacific island war. 

""Kessler had to train the flat-trajectory 
5"/51's so as to fire across Flipper Point and just 
clear the crest of Wake Island. The line of fire 
passed less than 230 yards to the west of Lt 
Kliewer's position at the generator. 

Meanwhile Second Lieutenant Robert 
W. Greeley had reached the command post 
with the first 20 men from Battery D. 
There Major Potter, trying to piece out 
and extend his sparse line to the right 
(west) , directed that the reinforcements be 
placed on that flank around the edge of the 
clearing originally dozed out to prepare 
for the north-south leg of the airstrip. 
Captain Godbold arrived with other re- 
inforcements at about 0700,^' and these 
men joined those already emplaced by 
Greeley. This line now turned to the right 
(north) to refuse the flank along the edge 
of the clearing. Potter's line, now con- 
taining about the equivalent of a rifle 
platoon, thus extended from near the 
beach, across the two roads south of the 
CP, and to the airstrip clearing whei'e it 
made a northward turn. Thus a gap of 
approximately 450 yards existed between 
the skirmish line and the shore of the 
lagoon. This gap the defenders would 
attempt to cover by fire. 

By daylight the atoll defenders could 
make out the large task force which sup- 
ported the landing operations. There 
were then 13 ships at various positions 
around the island (the four cruisers of 
Cruiser Division 6 were out of sight east 
of Wake), and all of them were keeping 
a safe distance from the 5-inch shore bat- 
teries. "Due to the previous experience 
with the American shore batteries," a 
senior Japanese officer said later, "we did 

" Like so many other questions as to exact 
times of events during the defense of Wake, this 
one is subject to conflicting testimony. Maj 
Potter states that Godbold reached the command 
post at 0600. Godbold gives 0715 as the time. 
Other sources, while not giving times, put the 
arrival of Battery D shortly after daybreak. 
Balancing all accounts against each other, 0700 
or shortly before seems to be the best synthesis. 



not want to come within range." ^® In 
spite of this caution, however, the de- 
stroyer Mutsuki began at 0654 to lead two 
other destroyers ( probably the Yayoi and 
the Mochizuki) in toward Wilkes Island, 
possibly to fire shore bombardment mis- 
sions. But fire from Battery B on Peale 
quickly hit the Mutsuki^ and the formation 
turned and scurried away. Observers be- 
lieved that Kessler's fire also hit the second 
destroyer in the formation after the ships 
turned, and that the Mutsuki later sank, 
but Japanese records do not confirm this. 

Farther to the northwest the two Japa- 
nese carriers Soryu and Hiryu headed up- 
wind with their cruiser and destroyer 
escort, ==» and at 0700 "the gallant Eagles 
of the Navy," as the Japanese Naval 
Information Service styled them, ap- 
proached Wake at 6,000 feet. As the for- 
mation wheeled over Peacock Point, Bat- 
tery E opened fire in what was the last 
antiaircraft action of the battle. The for- 
mation split into component groups ac- 
cording to mission, and commenced a 
methodical and unceasing series of air 
strikes in close support of the special 
landing force. Wilkes, Peale, and Wake 
Island were hit repeatedly. 

Dive bombers now battered Kessler's 
5-inch gun battery on Peale Island, and 
the air-supported enemy troops began to 
move rapidly against Major Potter's line 
south of the defense battalion command 
post. Battery E also Avas being attacked 
by the carrier planes, and Devereux be- 
lieved that Wilkes Island and most of the 

^ Hawaii-Malaya NavOps, 29. 

" In addition to the two carriers, this task 
force was composed of the new 12,000-ton heavy 
cruisers Chikuma and Tone, and six destroyers, 
two of which were Tanikaze and Vrakaze. 

west leg of Wake Island already had fal- 
len to the Japanese. Shortly after 0700 
the major called Commander Cunning- 
ham and told him that organized resist- 
ance could not last much longer. Was 
there a chance that the relief expedition 
might yet arrive ? No chance at all, Cun- 
ningham said.- 

And there was no chance, although up 
until two and a half hours earlier than 
this the men in Task Force 14 thought 
there might be. During the night of 22- 
23 December (21-22 December at Haw^aii) 
Vice Admiral William S. Pye, act- 
ing CinCPac pending the arrival of Ad- 
miral Nimitz from Washington, had been 
in conference about this relief force for 
Wake. The officers at Pearl Harbor knew 
that Admiral Fletcher was running a close 
race, and they were concerned that this 
task force would be lost, along with Wake, 
if the race ended in a dead heat. At one 
point they decided to order the Tangier to 
make a solitary dash for the atoll while 
the Saratoga, then some 425 miles short of 
Wake, launched Major McCaul's planes 
from that distance. But this order was 
countermanded before Fletcher could be- 
gin its execution; and finally at 0811 Ha- 
waiian time (some two and a half hours 
before Wake was to surrender) Task 
Force 14 was recalled. The force spent 
most of the day refueling its cruisers, and 
that night retired toward Midway. 

Commander Cunningham and Major 
Devereux decided that additional defense 
efforts would be hopeless, and the island 
commander made the decision to surren- 
der. Acting on these orders, Devereux 
carried a white flag out of his CP at 0730 
and walked south along the shore road to 
meet the Japanese. 




"At this time," states a Japanese report, 
"Wilkes Island was the scene of a fierce 
and desperate battle." ^^ Here at 0245 
Gunner McKinstry fired the first shots in 
the battle of Wake when he saw barges 
approaching to land at a point just west 
of the new channel. After this first burst 
from gun 10 (see Wilkes map) a Marine 
searchlight flashed on to reveal the 100 
men of the Takano unit coming through 
the surf and onto the beach. But by this 
time the Wilkes detachment was complete- 
ly disposed to repel a landing, thanks to 
the earlier and erroneous report of a land- 
ing on Peale Island. (See Map 6) 

When that false alarm sounded. Cap- 
tain Piatt ordered two Battery L gun sec- 
tions ( each about the size of a rifle squad ) 
to positions on the lagoon side of the is- 
land, and pulled the remainder of Battery 
L personnel back to defensive positions 
along the road near the new channel. Ex- 
tra ammunition and grenades were issued, 
and the Battery F personnel were in- 
structed to fire against any landing for as 
long as they could, and then to pull back 
across the road to join the men of Battery 
L. Thus the Battery L position, com- 
manded by Lieutenant McAlister, was 
well prepared when the Japanese barges 
hit the beach near that defensive site. 

The searchlight beam lasted for only 
a minute in the face of the Japanese at- 
tack, but McKinstry continued to fire at 
the landing craft he could see on the beach, 
and McAlister sent two men down toward 
the beach to hurl grenades at the Japanese. 
Enemy fire killed one of these men and 
wounded the other. Battery F then began 

'Hawaii-Malaya NavOps, 29. 

to fire into the landing area with their 
3-inch shells cut for muzzle-burst, but the 
attack came on up the beach so rapidly 
that these guns soon were unable to depress 
sufficiently to engage the Japanese. 

Gunner McKinstry had crossed to the 
3-inch guns to direct this air-burst fire, 
but it soon became apparent to him that 
the position could not hold. The enemy 
continued to expand their beachhead, and 
a strong force near Battery F was throw- 
ing grenades in among the Americans. 
The gunner removed the firing locks from 
the 3-inch guns, and then directed his men 
to retire to their designated infantry posi- 
tion on the right flank of McAlister's line 
beyond the road. Japanese tried to pursue 
this withdrawal, but McKinstry's men 
drove them back. 

McKinstry and McAlister now were in 
good position to protect themselves and to 
guard the road to Wake Island, but there 
was little to stop the Japanese from mov- 
ing farther west and spreading out over 
all of Wilkes Island unless fire from ma- 
chine-guns 9 and 10 could aid the main de- 
fense line to keep the enemy bottled up 
around the abandoned 3-inch guns. Gun 
9 was already delivering flanking fire 
against these Japanese, and the enemy 
advance was temporarily checked. The 
Takano troops now turned their attacks 
to knock out this machine gun, but its posi- 
tion was well prepared and well camou- 
flaged. Although nearly surrounded, the 
Marines on this gun continued to hold and 
to repel attacks which kept up until dawn. 

Meanwhile Captain Piatt, in his CP be- 
hind the former positions of Battery L, 
Avas having the same sort of communica- 
tions trouble that plagued Major Devereux 
on Wake Island. By 0300 the Captain had 
lost contact with every position except that 





of the beleaguered men on Gun 9. From 
them he learned that the enemy were build- 
ing up pressure to extend their beachhead 
farther inland. At about 0400 Captain 
Piatt moved out to the Gun 11 position 
near the beach, and from there he crept 
through the brush to a vantage point east 
of Gun 10. It was now about 0500, and 
Piatt decided quickly that he must mount 
a counterattack if the Japanese were to be 
prevented from staging daylight attacks 
which would enable them to overrun Gun 9 
and spread out into the interior of the 

He hurried back to Gun 10 and ordered 
Platoon Sergeant Raymond L. Coulson 
to round up the .30 caliber machine-gun 
crews and searchlight personnel from 
Kuku Point, plus anyone else he could 
lay hold of, and assemble them at Gun 10 
for the counterattack. In 25 minutes 
Coulson was back with the two machine- 
gun crews and eight riflemen — about a 
squad in all. These men the captain led 
back through the underbrush toward the 

The Marines crept and crawled to Avith- 
in 50 yards of the Japanese. Piatt then 
placed his two machine guns on each flank 
of his line of departure, and ordered the 
gunners to fire their short bursts close to 
the ground so this fire would not endanger 
the McAlister and McKinstry line farther 
to the east. By this time dawn was break- 
ing, and Piatt quickly drew up his skirmish 
line of eight Marines. He signalled the 
machine guns to open fire, and then he led 
his riflemen forward against the 100 men 
of tlie Takano xinit. 

At about this time on the other side of 
the Japanese position. Lieutenant McAl- 
ister had observed a six-man enemy patrol 
moving toward his Marines, and he or- 

dered his line to open fire. One enemy 
was killed and the others sought cover be- 
hind a large coral rock near the beach. 
McAlister's men continued to fire into 
this area to keep the Japanese pinned 
down while Gunner McKinstry and Pri- 
vate First Class William C. Halstead 
worked their way out to this rock and fin- 
ished off the rest of the patrol. 

Meanwhile Piatt's counterattack had 
surprised the other flank of the penetra- 
tion, and the Japanese at that point were 
in trouble. Obviously they had expected 
no opposition from the west, and their 
light machine guns had been sighted for 
fire to the east against the McAlister-Mc- 
Kinstry line. Piatt's attack carried the 
Marines into the former position of Bat- 
tery F, and the Japanese were driven back 
toward the beach and toward the Marine 
defense line by the island road. 

It was now daylight, and McAlister 
could see this Marine attack on the far 
side of the Japanese position. When his 
men finished mopping up the enemy 
around the rock near the beach, the lieu- 
tenant gathered 24 Marines into a skirmish 
line of his own and launched a counter- 
attack from his side of the battlefield. The 
men of the Takano landing force panicked. 
Organized resistance evaporated in front 
of the two Marine attacks, and the forces 
of Piatt and McAlister soon joined. 
About 30 Japanese fled to shelter around 
the Marine searchlight truck southeast of 
tlie Battery F guns, and there the Marines 
under Piatt and McAlister flushed them 
out and killed tliem. The Takano unit on 
Wilkes had been destroyed. 

McAlister counted four officer and 90 en- 
listed bodies while his men policed up the 
battlefield and removed the flags the Jap- 
anese had placed in the ground to mark' 



their front lines. Two wounded Japanese 
were captured. The other four Japanese — 
if the Takano tmit actually included an 
even 100 — were not accounted for. Ma- 
rines found several small maps of Wake 
in the effects of the dead Japanese, and 
Marine positions were marked accurately 
on these maps. The photographic mis- 
sions over the atoll had obviously paid off 

By 0800 Captain Piatt had reorganized 
his Wilkes defenders, and he again tried 
to establish contact with Wake Island. He 
was able to contact the motor pool at Camp 
One where Poindexter's force had man- 
aged to hold throughout the night, but 
he could not get through to Devereux at 
the defense battalion CP. At about noon 
the men on Wilkes observed Japanese land- 
ing boats headed for Wake Island and sev- 
eral ships approaching toward Wilkos 
channel. Piatt ordered McAlister to get 
his 5-inch guns into action against these 
vessels, but the gun crews found that the 
weapons were beyond use. The training- 
mechanism on Gun 1 was wrecked, and the 
Gun 2 recoil cylinder had been riddled by 
bomb fragments. 

Wilkes had been under attack by the dive 
bombers which had arrived over the atoll 
at about 0700, but sign language interroga- 
tion of the wounded prisoners indicated 
that the enemy planned no more landings 
against this section of the atoll. Piatt de- 
cided to go find the enemy. He ordered 
McAlister, McKinstry, and Coulson to 
round up all the men and to strike out east 
toward the old channel. Dive bombers 
attacked this route column as it moved 
down the island, and a destroyer moved 
in to open up from 2,000 yards. One Ma- 
rine, Private First Class Robert L. Ste- 
vens, was killed by this bombing, but the 

action against the Marines suddenly 

Piatt moved the men forward again in a 
dispersed formation, and near the old 
channel he saw three men advancing from 
the other direction. Two were obviously 
Marines, Piatt decided, but the figure in 
the rear was a Japanese officer armed with 
a large sword. The captain moved for- 
ward and soon recognized Major Devereux 
who told him that the island had been 
surrendered. It was then shortly after 
1330. Piatt's force did not get a chance 
to help in the fighting on Wake Island, but 
it had given such a good account of itself 
in earlier action that a Japanese officer was 
prompted later to make this estimate of 
the Wilkes fighting : "In general, that part 
of the operation was not successful." '^ 


Prior to moving down the road toward 
the Japanese, who were still receiving de- 
termined small-arms fire from the few Ma- 
rines south of the command post. Major 
Devereux passed word of the surrender to 
all units in communication with his com- 
mand post. These were Batteries A and 
E on Wake Island, Battery B on Peale, 
and other small detachments including 
those at Heel Point, and some of the .50 
caliber positions on Wake Island. Com- 
munications with Battery A had been re- 
stored at about daybreak. All units were 
ordered to destroy their materiel as best 
they could prior to actual surrender. 

These instructions were carried out with 
all possible thoroughness. At Battery E 
an attempt was made to damage the 3-inch 
antiaircraft guns by stuffing blankets into 
the muzzles and then firing a round or two. 
When this failed to produce appreciable 

"' Capture of Wake, II, 372. 



results, the firing locks were removed and 
smashed, and grenades were rolled down 
the muzzles to explode inside and damage 
the rifling. All electrical fire control data 
receivers were smashed, electric cables 
chopped up, and the battery commander 
fired twenty rounds of .45 caliber ammuni- 
tion through the delicate optical and elec- 
tro-mechanical parts of the height finder 
and director. After completing these 
measures. Lieutenant Lewis assembled the 
men of Battery E and marched them under 
a white flag to the battalion command 

At Battery A, the 5-inch firing locks 
were broken and buried, and all gun tele- 
scopes smashed. The range keeper was 
damaged beyond repair. After that a 
white flag was run up, and Lieutenant 
Barninger ordered his men to eat as much 
as they could hold. He then held his men 
on the position to await arrival of the Jap- 
anese. Elsewhere, the hard-pressed rifle- 
men stripped the bolts from their rifles and 
flung them into the brush. 

It was after 0800 before all this had been 
attended to, and the rifle fire of Potter's 
line was still covering the final operations 
of the command post. Major Devereux 
then tried to contact the Marine aid sta- 
tion located some 300 yards south of the 
CP. He believed that the Japanese ad- 
vance must have reached this point, and 
he wanted to instruct the battalion surgeon 
to contact the Japanese. But there was no 
response from the aid station, and it be- 
came apparent that a surrender party must 
go forward from the CP. Major Devereux 
and Sergeant Donald Malleck, who car- 
ried a white rag tied to a mop-handle, then 
made their way down the road toward the 
fighting. At the Marine line Devereux 
ordered Potter's men to hold their fire, and 

he and Malleck walked on toward the 

Near the hospital Devereux and the ser- 
geant were halted by a Japanese rifleman 
who motioned for them to throw down 
their arms and helmets. Then the soldier 
took them to the hospital where the Jap- 
anese already were in charge. They had 
killed one patient and wounded another 
while capturing the hospital, and now they 
had all the patients outside trussed up with 
telephone wire. Commander Cunning- 
liam arrived by truck while Devereux was 
explaining his mission to an English- 
speaking Japanese officer, and the Marine 
major turned over his surrender duties to 
the island commander. A Japanese officer 
then escorted Devereux and Malleck for- 
ward to pass the surrender order to Ma- 
rine units on the west leg of Wake Island 
and on Wilkes Island. 

They found the VMF-211 riflemen and 
Hanna's unit still holding around the 
3-inch gun in spite of continuing efforts 
by the Japanese. The Japanese, unable 
to advance, had taken up positions behind 
nearby plane revetments, and the fighting 
here was a deadlock. Captain Tharin was 
the only officer unwounded in the Marine 
position, and he was directing the action 
when Major Devereux contacted him at 
0930. There were now but 10 Marines sur- 
viving, and nine of them were wounded. 

At 1014 Devereux reached Lieutenant 
Kliewer and his three men beside the mine 
field generator. These men had been try- 
ing since 0900 to coax some life into the 
gasoline generator so they could blow up 
the airfield, but the rain during the night 
had given it a thorough soaking and it 
would not operate. "Don't surrender, 
lieutenant," one of the men told Kliewer. 
"Marines never surrender. It's a hoax." 



"It was a difficult thing to do," Kliewer 
reported later, "but we tore down our guns 
and turned ourselves over." ^^ 

Shortly before 1115 the surrender party, 
now west of the airstrip, came upon the 
rear of a Japanese skirmish line facing 
westward and evidently engaged in a fire 
fight against Marines in the brush beyond 
the west end of the strip. After some con- 
fusion during which the Japanese fired on 
the surrender group, Major Devereux 
passed through the lines and made contact 
with Lieutenant Poindexter. The lieu- 
tenant's mobile reserves, in ignorance of 
the surrender, had retaken the ground be- 
tween Camp One and the west end of the 
strip during the morning's fighting. When 
Devereux came upon Poindexter, the 3l)- 
odd Marines in this force had just com- 
pleted a steady eastward advance from 
Camp One, fighting their way forward 
along the beach with the edge of the brush 
to their left. Special naval landing force 
troops were in the thick brush to the north, 
but they had not attempted to attack the 
Marines. Divided into three 10-man 
squads, Poindexter's improvised platoon 
had advanced with two squads in assault, 
one on the seaward side of the road and 
the other north of the road. The support 
squad protected the exposed left flank by 
advancing in rear of the left assault squad. 

■ IstLt D. D. Kliewer Rept. 

During the advance, particularly as he 
neared the airfield and retraced by day- 
light the scenes of his fighting during the 
night. Lieutenant Poindexter counted ap- 
proximately 80 enemy dead. 

After assuring the surrender of this 
force. Major Devereux led the Japanese 
toward Camp One, still held by machine- 
gun sections of Poindexter's group. There 
the Marine prisoners watched a Japanese 
climb up the water tank and cut down the 
American flag which had been flying there 
throughout the battle. 

The surrender group, followed by ap- 
proximately 30 Japanese, then crossed 
Wilkes channel by launch. No Marines 
were to be seen when Devereux landed at 
about 1300, and the party began walking 
cautiously westward. At this time the 
enemy destroyer began firing on the island, 
but this fire was soon checked by a Jap- 
anese signalman who flagged the ship to 
silence. At 1330, almost midway between 
the new and old channels on Wilkes, Major 
Devereux saw "a few grubby, dirty men 
who came out of the brush with their rifles 
ready ..." These were Piatt's Marines 
who had annihilated the Takano landing 
party on Wilkes and now were advancing 
eastward to repel what they thought was 
still another landing. Thus all resistance 
had been silenced, and Wake now was in 
Japanese hands. 



The defense of "Wake was the first war- 
time operation conducted by the Marine 
Corps in defense of an advanced naval 
base. It was also the first combat test of 
the Marine defense battalion, although the 
strength of the Wake detachment was 
greatly reduced. The main reason for the 
fall of Wake seems obvious. The enemy in 
greatly superior strength, supported by 
ample surface and air forces, was able to 
effect a lodgement on the atoll and then to 
apply his ground superiority to over- 
whelm the dispersed defenders in detail. 
Had it been possible for U. S. surface 
forces to intervene, or for substantial re- 
inforcements to reach Wake, the results 
might have been entirely different. But 
military lessons of some value still may be 
drawn by a survey of certain specific rea- 
sons why the defense was handicapped. 
These factors were interacting, of course. 
No single one of them can be clearly iso- 
lated within the framework of events 
which brought military defeat to the atoll. 

Japanese procedure for the reduction 
and seizure of Wake, if not executed with 
the skill or standards that U. S. forces 
later attained, was nevertheless orthodox. 
It consisted essentially of two phases, the 
preliminary bombardment and the assault 
landing. The enemy's first landing plan 
underestimated the amount of preparation 
required, and he paid for this miscalcula- 
tion in the defeat of 11 December. But 
this he corrected in his second attempt. 

Lack of radar and other early-warning 
equipment severely handicapped Marines 
during preliminary aerial bombardment, 
and it would be difficult to overstate the 
seriousness of this shortage. It enabled 
the initial Japanese raid to destroy over 
half of VMF-211's fighters on the ground, 
and the same lack of early warning con- 
tinued to hamper the effectiveness of those 
fighter planes which remained in opera- 
tion. Thus the VMF-211 pilots never had 
a chance to plan effective fighter intiercep- 
tion against the enemy bombers, and the 
Japanese could proceed quite methodically 
with their program for the aerial soften- 
ing of Wake. 

This lack of early warning and the 
shortage of aircraft can be lumped to- 
gether as matters of air defense, and air 
defense depends upon coordinated em- 
ployment of fighter aircraft, antiaircraft 
artillery, and the essential warning sys- 
tems. But on Wake only the antiaircraft 
artillery — undermanned and partially op- 
erational though it was — could be consid- 
ered fully and consistently effective, and 
nobody ever expected antiaircraft weapons 
alone to defend an advanced naval base 
against air attack. They were there to 
provide close-in protection to the aviation 
facilities; the planes were to be the im- 
portant factor in keeping the enemy away 
from an island base. Determination and 
stubbornness of the fighter pilots could not 
avert the final outcome. The fliers could 
only exact from the enemy a maximum 
cost for every bomb dropped. This was 




done until the last Grumman was de- 
stroyed by massed enemy fighters on 22 
December. After tliat, landing opera- 
tions against Wake could proceed. 

Once the ground combat began, the fun- 
damental weakness of the defense bat- 
talion concept as it then existed became 
starkly underlined. The unit had no in- 
fantry component to act as an effective 
mobile reserve. Most garrison personnel 
were tied to weapons and battery positions, 
and Major Devereux could muster only a 
fraction of his manpower against the in- 
vaders even after enemy intentions became 
apparent. On Wake Island, for example, 
only about 85 of 200 Marines were readily 
available to check the assault landing of 
a thousand Japanese. Militarily speak- 
ing, there is something pathetic in the 
spectacle of Lieutenant Poindexter and his 
"mobile reserve" of eight men and four 
machine guns dashing by truck from one 
threatened point to another in the face of 
such fantastic odds. 

True, at that time trained infantry was 
almost as scarce as radar. But the fault 
lay in the defense battalion tables of or- 
ganization. Later this omission was cor- 
rected, and Midway had both infantry and 
light tanks. Had even one Marine in- 
fantry company reinforced with tanks 
been on Wake, it is possible that the gar- 
rison might have thrown the Japanese 
back into the sea. This is borne out by 
what happened on Wilkes Island, where 
Captain Piatt was able to annihilate twice 
his numbers of the enemy by shrewd, co- 
ordinated counter-attack. And after day- 
light on Wake Island, Poindexter, with 
the makeshift defenders of Camp One 
added to his "mobile reserve," had as- 
sumed the offensive, driven back the Jap- 
anese to his front, and regained most of 

the ground given up during the confused 
hours of darkness. 

After the Japanese had landed in force 
on the south coast of Wake Island, it ap- 
pears that the coast artillery and antiair- 
craft missions of Batteries B and D, re- 
spectively, had become of secondary im- 
portance in light of the serious enemy 
ground threat. The military reader might 
wonder why all available personnel from 
Batteries B and D, with whom Devereux 
was still in communication, were not early 
in the battle brought down to the vicinity 
of the airfield and employed, together with 
such few other available Marines, as a mo- 
bile reserve to counterattack the main Jap- 
anese beachhead. This was partially ac- 
complished at 0530 on the final morning 
when Captain Godbold was directed to 
bring the personnel of his battery (D) 
to the command post for employment as 
infantry. By this time, however, it was 
too late for such a small number to influ- 
ence the outcome of the battle. In this 
connection. Major Devereux later pointed 
out that because of the partial failure of 
communications he never had anything 
like a clear picture of the situation during 
the final Japanese attack. For several 
hours he was in doubt as to the location of 
the main enemy landing and hence did not 
consider himself justified in stripping 
Peale Island of all defenders. 

As alluded to above, another major les- 
son to be derived from this phase of the 
operation was a re-emphasis of Admiral 
Mahan's famous dictum that "Communi- 
cations dominate war." The partial fail- 
ure of communications, which occurred 
shortly after the Japanese landing, isolat- 
ed the defense detachment commander 
from most of his subordinate units then in 
action. As a result he not only lost control 



over much of the battle, but he also — and 
perhaps more important to this case — be- 
came unavoidably deceived as to the prog- 
ress of the situation. In ignorance of 
what happened on Wilkes or at Camp One, 
he surmised that all was lost in those areas. 
Buried telephone lines and reliable field 
radios would have prevented this failure 
of communication, and the surrender de- 
cision would not have been made at that 
particular stage of the action. The Wake 
garrison, however, had neither the per- 
sonnel to dig by hand, nor the machinery 
to dig by mechanical means, the many 
miles of ditches which would have been 
necessary to bury the telephone lines. 

Perhaps one of the fundamental reasons 
for the state of the Wake defenses 
stemmed from the fact that base develop- 
ment had consistently received priority 
over defense preparations. That the de- 
fensive installations were in as good a con- 
dition as they were when the Japanese 
struck may be credited to the tremendous 
efforts of the small Marine garrison. 

All things taken into account, however, 
the decision to surrender Wake was rea- 
sonable, especially when considered in 
light of the civilian situation and the fact 
that relief was no longer in prospect. Ma- 
rines who fought through the Pacific cam- 
paigns would later see many examples of 
a totally unreasoning enemy who never 
surrendered but was always defeated. At 
the same time, insensibly, some might come 
to believe that unyielding refusal to sur- 
render was the proper role of a defender. 
Of course this was neither true nor logical. 
Wake had exacted a full and more than 
honorable toll from the Japanese, but its 
defensive resources had been exhausted. 

No fighter aircraft remained. Only one 
antiaircraft battery was effectively opera- 

tional. Enemy dive bombers on 23 De- 
cember had completely disabled one 5-inch 
battery (Wilkes) and largely destroyed 
the fire control instruments of the remain- 
ing two. Without airplanes, fire control 
instruments, radar, spare parts, and per- 
sonnel to bring the defense to full strength 
Wake could not carry on. The only 
answer was surrender. This took place 
fifteen days after the initial attack, and it 
was eleven hours after the fighting com- 
menced on shore before Wilkes Island sur- 

During this period the Marines sus- 
tained almost, 20 per cent casualties, but 
they exacted a heavy toll from the Jap- 
anese. Nearly 500 enemy had been lost 
in the abortive landing attempt of 11 De- 
cember, the defenders on Wilkes Island 
accounted for nearly 100 in their defeat of 
the Takrmo unit, and Poindexter counted 
approximately 80 enemy bodies during his 
morning attack from Camp One. Give the 
Hanna-VMF-211 position credit for at 
least 20 more kills, and this would bring 
the Wake total to 700 enemy. Others must 
have lost their lives on Wake Island land- 
ing beaches and elsewhere on the island, 
although the figure probably would not be 
great. But in earlier action the atoll anti- 
aircraft and fighter plane fire had downed 
21 enemy aircraft and claimed credit for 
damaging another 11.^ 

Based on this record, Major Putnam's 
final VMF-211 report of 21 December 
would truthfully state that "All hands 
have behaved splendidly and held up in 
a manner of which the Marine Corps may 
well tell." 

^ A Japanese source says that 51 planes, in 
addition to those shot down, were damaged by 
flak over Wake. Nakwnmra Notebook. 


Marines in the Philippines 

China and Luzon 


In the first few months after Pearl Har- 
bor, it seemed that nothing could stop the 
Japanese. One by one, the western out- 
posts in the Far East were overwhelmed. 
Allied ground troops, in desperately un- 
equal contests, were forced to retreat, fight, 
and retreat again; at sea and in the air 
the pitifully few ships and planes which 
had survived the initial onslaught were 
hoarded against the surety of further en- 
emy advances. A grim holding battle was 
joined along a line protecting Australia 
and New Zealand and their South Pacific 
lifeline to the States. Yet, despite its 
strategic importance, this vital defensive 
action gave first place in the news to the 
outcome of a hopeless struggle hundreds of 
miles behind the enemy's forward posi- 

For almost five months, two names — 
Bataan and Corregidor — dominated the 
headlines, taking fire in the minds of the 
Allied peoples as symbols of courage and 
devotion to duty. To the Japanese, who 
realized that they could starve out the em- 
battled defenders at little cost to them- 
selves, it became imperative that the issue 
be decided forthwith in battle. On the 
eve of the all-out offensive that brought the 
end on Bataan, the Japanese commander, 
addressing his combat leaders, clearly 
stated the importance of the isolated strong 
points in the eyes of the world : 

The operations in the Bataan Peninsula and 
the Corregidor Fortress are not merely a local 
operation of the Great East Asia War. This 

battle has lasted for about three months as com- 
pared with our speedy victories in Malaya, Dutch 
East Indies, and other areas in the Philippines. 
As the Anti-Axis powers propagandize about this 
battle as being a uniquely hopeful battle and the 
first step toward eventual victory, the rest of 
the world has concentrated upon the progress of 
the battle tactics on this small peninsula. Hence, 
the victories of these operations do not only mean 
the suppression of the Philippines, but will also 
have a bearing upon the English and Americans 
and their attitude toward continuing the war.^ 

Lieutenant General Masaharu Homma 
was right : the outcome of the battle did 
have a direct bearing on the Allied atti- 
tude toward vigorous pursuit of the war. 
Perhaps in no instance since the defense 
of the Alamo stirred Americans in another 
century did an unsuccessful battle carrj' 
within its waging and its ending the source 
of so much national pride and dedication. 


On 26 July 1941, shortly after Japan oc- 
cupied military bases in Indo-China, Pres- 
ident Roosevelt authorized the mobiliza- 
tion of the Philippine Army. The War 

^ HistSec, G-2, GHQ, FEC, Japanese Studies in 
WWII No. 1, 14th Army Ops, 2 vols., n. d. (lo- 
cated at OCMH), 141-142, hereinafter cited as 
IJith Army Rept. 

^ Unless otherwise noted the material in 
this section is derived from Adm T. C. Hart, 
Narrative of Events, AsFlt Leading up to War 
and From 8Dee41 to 5Feb42, written before 
llJun42 (located at NHD), hereinafter cited as 
Hart Narrative; Adm T. C. Hart, Supplementary 
Narrative to Hart Narrative, 80ct46 (located at 
NHD) ; Gen J. C. Wainwright, Rept of Ops of 




Department, which had requested this 
move, followed through with a directive 
organizing a new command, USAFFE 
(United States Army Forces in the Far 
East) , which included all American Army 
and Commonwealth troops in the Philip- 
pines. To head USAFFE the Army 
called out of retirement its former chief of 
staff, General MacArthur, who had served 
as Military Advisor to the Commonwealth 
Government since 1935. He was given 
rank as a lieutenant general and with char- 
acteristic energy tackled the enormous job 
of putting the Philippines into a state of 
readiness against attack. 

The bulk of USAFFE's troop strength 
was drawn from the Philippine Army 
which was, in July 1941, an army in name 
only. It consisted of the islands' police 
force, the 6,000-man Philippine Constab- 
ulary, a token air force and inshore naval 
patrol, and ten territorial reserve divis- 
ions. Since the start of the Common- 
wealth's defense training program in 1936 
about 110,000 Filipinos had received a few 
months of basic military instruction, but 
most of these reservists had no experience 
with crew-served weapons and only rudi- 
mentary knowledge of their own pieces. 

USAFFE and USFIP in the Philippine Islands 
1941-42, 10Aug46 (located at TAGO), herein- 
after cited as VSAFFE-VSFIP Rept; Annex 
VIII to USAFFE-USFIP Rept, MajGen G. F. 
Moore, Rept of CA CJomd and the Harbor De- 
fenses of Manila and Subic Bay, 19Feb41- 
6May42, 15Dec45 (located at TAGO), hereinafter 
cited as Moore Rept; BriGen S. L. Howard, "Re- 
port of the operation, employment and supply of 
the old 4th Marines from September, 1941 to the 
surrender of Corregidor, May 6, 1942," 26Sep45, 
hereinafter cited as Howard Rept; IJfth Army 
Rept; L. Morton, The Fall of the Philippines — 
United States Army in World War II (Washing- 
ton : OCMH, DA, 1953), hereinafter cited as Fall 
of the Philippines ; Rising Sun in the Pacific. 

The divisions had never operated as such 
in field maneuvers and were scantily pro- 
vided with arms and equipment. In order 
to mold an effective fighting force from 
the Philippine Army, MacArthur needed 
just about everything in the military sup- 
ply catalogs, but most of all he needed 
time — time for training, time for materiel 
and men to reach the Philippines from the 
United States. 

The instructors and cadres needed for 
training the Philippine Army were drawn 
from the Constabulary and the regular 
Army units available to USAFFE. Most 
of the 22,000 U. S. Army troops in the 
islands were serving in Coast Artillery 
regiments, the Army Air Corps, or the 
Philippine Division, sole regular infantry 
division in the islands. Over half of these 
men were members of crack Philippine 
Scout units.^ The regulars suffered, too, 
from a general lack of up-to-date weapons 
and equipment,* but they were well trained 
to use what they had. 

The War Department supported Mac- 
Arthur's requests for additional troops 
and supplies to the fullest extent possible 
in light of the country's world-wide com- 
mitments ; USAFFE received priority in 
almost every man power and materiel 
category. More than 7,000 men, mostly 
members of service and air units, and the 

' The Philippine Scouts was a U. S. Army 
organization in which the enlisted men were 
native Filipinos and most of the officers were 
Americans. The Scouts had and merited a high 
reputation for fielding units with good morale, 
excellent discipline, and a consistently superior 
level of combat readiness. 

* At the outbreak of the war, "the Philip- 
pine Division, less than two-thirds strength, had 
only three (3) new 37mm automatic firing can- 
non, three (3) 81mm mortars per infantry regi- 
ment and no (0) 60mm mortars . . ." 



major portion of the United States' heavj' 
bomber strength reached the Philippines 
prior to the outbreak of war. Much more 
was promised and planned, but the Jap- 
anese surprise attack effectively cut off the 
flow of reinforcement. It also forced a re- 
vision of MacArthur's defensive strategj'. 

In view of his healthy reinforcement 
prospects, the USAFFE commander had 
adopted an aggressive defense plan that 
conceded the enemy nothing. He did not 
expect the Japanese to attack before April 
1942 ^ and by that time he considered that 
his air and ground strength would be such 
that he could successfully hold his posi- 
tion against any attacking force. He was 
confident that the Philippine Army, when 
adequately trained and equipped, would 
be a match for the Japanese. 

The Commander in Chief of the U. S. 
Navy's Asiatic Fleet (CinCAF), Admiral 
Thomas C. Hart, was in substantial agree- 
ment with MacArthur's philosophy of an 
aggressive defense. He recommended that 
in the event of war his fleet units remain 
based at Manila Bay and fight the Jap- 
anese in Philippine waters. The Navy 
Department, however, adhered to its long- 
established plan that the major ships of 
the fleet would retire to the south at the 
imminence of war, to a base of operations 
in the Netherlands East Indies or Malaya, 
where they could cooperate with Allied 
naval units.® Hart's slim collection of 

^Gen J. C. Wainwright, General Wainwright's 
Story, R. Considine, ed. (Garden City, N. T. : 
Doubleday and Company, Inc., 1946), 13, here- 
inafter cited as Wainwright's Story. 

' After the war, in supplementary comments to 
his original report, Adm Hart agreed that the 
Navy Dept decision was the best that could have 
been made considering the situation at the time. 
He added, however, that if his original proposal 
of 270ct41 to continue to base at Manila had been 

cruisers, destroyers, submarines, and aux- 
iliaries, was certainly no match for the 
Japanese fleet, nor was it intended to be. 
The U. S. Pacific Fleet, based at Pearl 
Harbor, was the American striking force, 
and war plans envisaged its fighting ad- 
vance to the Philippines if the Japanese 

The Asiatic Fleet's major shore installa- 
tions were located at Olongapo on Subic 
Bay and at Mariveles and Cavite within 
Manila Bay. Since denial of Manila Bay 
to the enemy was a key point in war plan- 
ning, the activities of the 16th Naval 
District (Rear Admiral Francis W. Rock- 
well), the shore establishment supporting 
the Asiatic Fleet, were closely coordinated 
with USAFFE's defensive preparations. 
Contact mines were laid to connect with 
^controlled mine fields of the Army's har- 
bor defenses, completely closing Manila 
Bay. On Corregidor, site of the prospec- 
tive command post for the defense of 
Luzon, protected installations for naval 
headquarters, a radio intercept station, and 
a torpedo replenishment depot "jvere pre- 
pared and equipped. Large quantities of 
fuel and ammunition stored at Cavite were 
moved to dumps away from the naval base 
to lessen their vulnerability to bombing. 
(See Maps 7 and 8, Map Section) 

If the Japanese attacked, the most dan- 
gerously exposed elements of the Asiatic 
Fleet were those stationed in China : seven 
Yangtze River gunboats ; Colonel Samuel 
L. Howard's 4th Marine Regiment at 
Shanghai ; and the Marine embassy guard 
detachments at Peiping and Tientsin. 

turned down sooner he might well have made a 
better disposition of his fleet units ; he had acted 
on the assumption that his proposal would be 
accepted until it was disapproved in late No- 



Admiral Hart had begun making informal 
proposals that his China forces be with- 
drawn early in 1941, and after July, when 
he was "entirely convinced that the war 
was coming,'' ^ he followed up with em- 
phatic official recommendations that his 
men be gotten out before it was too late. 
Japanese war preparations were so evident 
by 1 September that the American Consul- 
General at Shanghai, the commander of 
the Yangtze Patrol, and Colonel Howard 
jointly recommended that all naval forces 
in China be withdrawn. Hart naturally 
concurred and further recommended to the 
Navy Department that the troops be evacu- 
ated in late September when the transport 
Henderson made a routine call on Chinese 
ports to pick up short-timers and other 

Hart's request was turned down as far 
as withdrawal on the Henderson was con- 
cerned. He was told, however, that joint 
State-Navy conferences would be held 
within a couple of weeks time to consider 
the problem of a withdrawal and its 
effect on negotiations for a settlement of 
Japanese-American diflFerences. Despite 
CinCAF's protest that this "was not a 
question that could be delayed for weeks 
but must be acted upon immediately," ® 
he did not receive permission to withdraw 
the gunboats and the Marines until 10 No- 
vember, "embarrassingly late" as he later 
noted.® Five of the gunboats were able to 
reach Manila without hinderance once 
clearance to leave was given.^" 

' Adm T. C. Hart Itr to CMC, 10Oct56, herein- 
after cited as Hart Comments. 

" Quoted in Howard Rept, 1. 

" Hart Narrative, 29. 

" The smallest of the gunboats, the Wake, was 
stripped and left at Shanghai to be used as a sta- 
tion ship and radio outlet for the remaining 

Two President liners, the Madison and 
the Harrison, were chartered to transport 
the Marines, attached naval personnel, and 
their supplies and equipment; provision 
was also made to evacuate some American 
civilians from Shanghai on the same ships. 
After it reached the Philippines and un- 
loaded, the Harrison was to return to 
North China and pick up the embassy 
guards and their gear at Chinwangtao. 
All signs pointed to the necessity for haste 
in the withdrawal. 

The Japanese were replacing their sea- 
soned troops around Shanghai with re- 
cruits, and large numbers of special 
armored landing barges which had previ- 
ously been seen near the city disappeared ; 
intelligence pointed to movement south- 
ward of both veteran units and landing 
craft. Intelligence also indicated that the 
Japanese Army was eager to take over the 
International Settlement, by force if nec- 
essary, and that it was only being re- 
strained by the Nipponese Navy's desire 
for an "incident" which would seem to 
justify such action. Several attempts were 
made to manufacture incidents, but the 
Marines refused to knuckle under to the 
pressure, and Colonel Howard initiated 
prompt action which kept the American 
defense sector clear. A copy of a Japanese 
warning order was obtained which stated 
that "in the event of war the 4th Marines 
would attempt to break through [our] 
lines;" " ample evidence of this belief was 
seen in the increase in size and number of 
the patrols in the city and in the construc- 

Americans ; it was captured on 8 December. The 
Tutuila at Chungking was turned over to the 
Chinese Nationalist Government under lend-lease 
since it could not get downstream through the 
Japanese blockade. 
" Howard Rept, 3. 



tion of concrete blockhouses on all roads 
leading out of Shanghai. 

Both the Madison and Harrison needed 
to be converted to troop use after their 
arrival at Shanghai, and the first ship was 
not ready until 27 November. By 1600 
that date, the Madison with half the regi- 
ment and half its equipment on board 
sailed for Olongapo. While this forward 
echelon, the 2d Battalion and half of the 
Kegimental Headquarters and Service 
Companies, was loading out, a message was 
received from CinCAF to expedite the 
evacuation. Even though the conversion 
work on the Harrison was three days short 
of completion, the decision was made to 
clear Shanghai the following day with the 
rest of the regiment and its remaining 

Despite the short notice and the harass- 
ing tactics of the Japanese, ^^ the Eegi- 
mental R-4 and Quartermaster, Major 
Reginald H. Ridgely, Jr., was able to load 
all organizational gear, over 500 tons, by 
1300 on the 28th. At 0900 that morning, 
the regiment assembled at the 1st Batta- 
lion's billet, formed up behind its band, 
and marched down Bubbling Well-Nank- 
ing Roads to the President Line's dock on 
the Bund. Thousands of cheering people 
lined the route of march, and the banks of 

" "All supplies had to pass through the Japa- 
nese Sector on the way to the Customs dock. 
About 3 : 00 p.m. November 27th they closed 
the Garden Bridge over Soochow Creek to traffic 
and our trucks were delayed nearly an hour 
before Contact could be made with the Japanese 
Admiral to get this bridge reopened to traffic. 
Customs officials ostensibly at the instigation 
of the Japanese were insistent that our supplies 
pass through the Custom House, but we ignored 
such orders and loaded them on lighters. The 
Japanese instigated three strikes during the 
night by the laborers loading the lighters." 
lUd., 4. 

the river were alive with flag-waving 
Chinese as a power lighter took the Ma- 
rines downstream to their ship. At 1400 
the Harrison weighed anchor and sailed 
for the Philippines, marking the end of a 
colorful era in Marine annals. 

As soon as the Harrison cleared the 
Whangpoo River, machine guns were 
broken out and manned for antiaircraft 
defense, and blackout regulations were put 
into effect.^^ Flights of Japanese aircraft 
checked the liner regularly as it moved 
out into the China Sea, but there were no 
incidents, and contact was made on the 
29th with submarine escorts dispatched by 
Admiral Hart. On 30 November and 1 
December the two transports arrived at 
Olongapo where the troops disembarked. 
Only a few supplies were unloaded at the 
naval station, ostensively because CinCAF 
had issued orders that the ships must pass 
through the mine field into Manila Bay by 
nightfall on the day of arrival. Actually, 
Admiral Hart had given oral orders to his 
staff that the Marines were to be landed 
with field equipment only, because it was 
his intention that : 

. . . they would get into the field, near Olon- 
gapo, as soon as they could. We [Hart and his 
staff] all knew that they had been cooped up in 
Shanghai through all those years where condi- 
tions for any sort of field training were very 
poor — and we thought that not much time re- 

While the regiment's heavy equipment 
was unloaded at Manila and trucked to 
Olongapo, the Harrison was readied for 
a return voyage to pick up the Marines 
from Peiping and Tientsin. It was al- 

" No special defensive arrangements were 
m.ade by the 4th Mar elements on the President 
Madison. CWO C. R. Jackson Itr to CMC, 
10Oct56, hereinafter cited as Jackson. 

^^ Hart Comments. 



ready too late, however, to rescue the 
North China Marines. The Japanese war 
plans liad been activated, and the carrier 
task force that would strike Pearl Harbor 
was at sea en route to its target. The 
troops, ships, and planes that would be 
sent against the Philippines were concen- 
trated at Formosa, the Ryukyus, and the 
Palaus with orders to begin their attack 
on X-Day — 8 December 1941 (Manila 
Time) .^^ 

6' DECEMBER 19 kV^ 

When the dawn of the first day of the 
Pacific War reached the China Coast, the 
attack on Pearl Harbor was over and the 
troops in the Philippines had been alerted 
to their danger. At the Chinwangtao 
docks. Second Lieutenant Richard M. Hui- 
zenga was supervising the stockpiling of 
supplies for the expected arrival of the 
President Harrison. A truck driver 
brought him word that the radio at his 
railhead, Camp Holcomb, was full of news 
of Pearl Harbor. Although the Japanese 
made half-hearted attempts to stop him on 
his three-mile drive back to the camp, 
Huizenga was able to get through to his 
unit. He found the 21 Marines of the 
loading detail surrounded, at a respectful 
distance, by a cordon of Japanese troops. 
The men, under Chief Marine Gunner 
William A. Lee, were setting up a strong 
point amid the boxcars of supplies; two 

" Campaigns of the Pacific War, 26-27, 50. 

" Unless otherwise noted the material in this 
section is derived from Hart Narrative; 
V8AFFE-U8FIP Rept; llfth Army Rept; How- 
ard Rept; 4th Mar Jnl and Rec of Events, 
8Dec41-2May42, hereinafter cited as ^th Mar 
Jnl; Capt A. F. Metze Rept to CMC, "Surren- 
der of U. S. Marine Forces in North China," 
23Aug42; Fall of the Philippines; Rising Sun in 
the Pacific. 

machine guns and several Tommy guns 
and BARs had already been broken out 
of their cosmoline packing. Despite their 
desperate situation the Marines were 
ready to fight. 

Huizenga and a Japanese captain held 
an armed parley where the lieutenant was 
given time to communicate to his superior 
at Tientsin, Major Luther A. Brown, the 
enemy's demand that he surrender the de- 
tachment. Orders soon came back to offer 
no resistance and the Marines were 
stripped of their weapons. Later in the 
day they were returned under Japanese 
guard to the Marine barracks at Tient- 
sin. ^^ 

The situation of the detachments at 
Tientsin and Peiping was similar to that 
of the one at Camp Holcomb; Japanese 
troops surrounded their barracks in 
strength and demanded their surrender. 
Since the embassy guard was not required 
to maintain a continuous watch on Cin- 
CAF's command radio circuit, ^* the first 
word that the senior Marine officer. Colo- 
nel William W. Ashurst, had of the out- 
break of hostilities came from the Japa- 
nese. He was given till noon to make his 
decision whether to fight or not and was 
allowed to communicate by radio with 
CinCAF and by phone with Major 
Brown. In a sense Ashurst had been given 
a Hobson's choice : he could surrender or 
he could let his troops, fewer than 200 of- 
ficers and men, be overwhelmed. If dis- 
cipline and spirit would have won the day. 

"MIS, G-2, WD, Escape Rept No. 665, Capt 
R. M. Huizenga, 12Jul45. 

"'LtCol W. T. Clement Rept to CMC, "Dis- 
positions and employment of U. S. Marines on 
the Asiatic Station during the initial stages of 
the war," 6Apr42, hereinafter cited as Clement 



Ashurst could have opened fire on the be- 
siegers — his men had already demon- 
strated at Camp Holcomb that they were 
willing to take on hopeless odds. But 
there was no purpose in fighting if the end 
result could only be useless bloodshed. 

In the absence of instructions to the con- 
trary, Colonel Ashurst took the only sen- 
sible course open to him and ordered his 
men to lay down their arms. A strong 
possibility existed that if no resistance 
was offered the embassy guards would be 
considered part of the diplomatic entou- 
rage, entitled to repatriation. As the 
initial treatment of the Marines was rela- 
tively mild and they repeatedly received 
informal assurances from the Japanese 
that they would be exchanged, few at- 
tempted escape. When these rumors 
proved false, the opportunity had 

By the time Ashurst's report of his de- 
cision to surrender reached Hart in Ma- 
nila, the Philippines were in the thick of 
the war. The first news of the Japanese 
attack was picked up at 0257 by a radio 
operator at CinCAF Headquarters who, 
recognizing the technique of the sender, 
vouched for the reliability of the now 
famous message, "Air raid on Pearl Har- 
bor. This is no drill." ^^ The duty of- 
ficer, Marine Lieutenant Colonel William 
T. Clement of Hart's staff, immediately 
notified the admiral who sent a war alert 
to all fleet units. Minutes later, by a com- 
bination of intercepted official and com- 
mercial broadcasts and the spreading of 
the word by the first agencies notified, the 
report had reached all major USAFFE 

A cacophony of sound broke the still- 
ness at Olongapo when the alert reached 
the naval base at 0360; the bugler of the 
guard blew "Call to Arms;" the steam 
whistle at the power plant blasted a recall 
signal to PBY crewmen; and the ship's 
bell at the main gate clanged continu- 
ously.^^ Companies immediately mus- 
tered in front of their wooden barracks 
and in the streets of tent areas and were 
put to work setting up machine guns for 
antiaircraft defense and digging individ- 
ual protective holes. Colonel Howard 
initiated the first moves in what was to be 
a hectic period of redisposing, reorgan- 
izing, and reinforcing the regiment which 
lasted throughout the month of December. 

When the 4th Marines arrived from 
Shanghai its strength stood at 44 officers 
and warrant officers and 728 enlisted men ; 
organic naval medical personnel raised the 
total strength to 804. The regiment "had 
been permitted to dwindle by attrition" ^^ 
in China so that it consisted only of Head- 
quarters Company, Service Company, and 
two battalions — the battalions short one of 
their rifle companies and the companies 
each short one of their three rifle platoons. 
By utilizing the members of the regimen- 
tal band and absorbing the Marine Bar- 
racks Detachment, Olongapo, Howard 
was able to form some of the missing pla- 

"Huizenga, op. oit.; MIS, G-2, WD, Escape 
Rept No. 666, Capt J. D. McBrayer, Jr., 12Jul45. 
"Clement Rept. 

'^ Capt F. W. Ferguson, Personal Experiences 
8Dec41-6May42, n. d., hereinafter cited as Fer- 

^ Howard Rept, 8. As the threat of war with 
Japan increased, Adm Hart initiated a policy of 
withholding replacements from Marine units in 
China. Almost all of the men held back were 
assigned to the 1st SepMarBn at Cavite. Hart 
felt that if by some mischance he was unable 
to get the 4th Mar out of China he "could at 
least stop sending any more Marines there until 
someone bawled us out most vociferously. They 
never did." Hart Comments. 



toons. In keeping with previous orders 
from Admiral Rockwell, he sent the 1st 
Battalion by tug, lighter, and truck to 
Mariveles to relieve the Marine detach- 
ment there. The men of the Mariveles 
guard had been taken from the other ma- 
jor Marine unit in the Philippines, the 1st 
Separate Marine Battalion at Cavite. 

The battalion was organized to function 
either as an antiaircraft or an infantry 
unit, but its primary mission was the anti- 
aircraft defense of the naval installations 
in the Cavite-Sangley Point area. Its 
firing batteries, 3-inch guns and .50 caliber 
machine guns, had been on partial alert 
since 14 October and as the threat of war 
grew stronger the guns and their crews 
had reached a high degree of readiness. 
On 4 December, the battalion's one long 
range radar set and the necessary operat- 
ing personnel were assigned to 
USAFFE's control and moved to a posi- 
tion on the west coast where the radar 
could scan the approaches to Manila from 
the south ; ^^ the set was one of two operat- 
ing in the Philippines on 8 December. 
When the battalion commander. Lieuten- 
ant Colonel John P. Adams, passed the 
word of the Pearl Harbor attack there was 
little left to be done but to "cut fuzes, 
going into the last stage of readiness." ^* 
But the Marine guns were not to see ac- 
tion. The Japanese reserved their first 
day of attack for their primary target, 
MacArthur's Far East Air Force 
(FEAF). And when that day was over, 
the Japanese figured that their "suprem- 

acy, at least in the air, was established 
conclusively." ^^ 

Except for 16 B-l7's at Del Monte on 
Mindanao, the bulk of FEAF's strength 
in first-line planes was stationed at fields 
in central Luzon. Dawn of the 8th found 
most of these planes airborne, waiting to 
engage or evade Japanese attackers. But 
the land-based naval fighters and bombers 
of the Japanese Eleventh Air Fleet, 
charged with making the main air assault, 
did not appear at dawn. The enemy plan 
had called for such a surprise attack, 
timed to coincide with the start of opera- 
tions in Malaya and at Pearl Harbor, but 
thick clouds and heavy fog delayed the 
take-off from Formosa of the major attack 
formations. It was noon before the enemy 
planes could reach their targets, Luzon's 
airfields, and the Japanese pilots very 
reasonably assumed that with the loss of 
surprise they would be met in force.^* 

But this was not to be, for "shortly 
after 1130 all American aircraft in the 
Philippines, with the exception of one or 
two planes, were on the ground." ^^ The 
fighters were refueling after their fruit- 
less morning^ patrols or awaiting a warn- 
ing of imminent attack ; the bombers were 
arming for an offensive mission against 
Formosa. By an incredible chain of cir- 
cumstances, compounded by poor com- 
munications, a woefully inadequate air 
warning system, and a generous amount 

"^ LtCol H. L. Davis Itr to CMC, 310ct56. 

^ IstLt W. F. Hogaboom, Personal Experiences 
140ct41-6May42, n. d., 1-2, hereinafter cited as 
Hogaboom. This narrative was published in the 
MC Gazette, April 1946, under the title, "Action 
Report — Bataan." 

'^ HistSec, Japanese Research Div, GHQ, EEC, 
Japanese Studies in WWII No. 11, Philippines 
AirOpsRec, lFeb52, 7, hereinafter cited as 
Philippines AirOpsRec. 

"-" Ibid. 

" W. S. Craven and J. L. Gate (eds.). Plans and 
Early Operations, January 1939 to August 
1942 — The Army Air Forces in World War II 
(Chicago: University of Chicago Press, 1948), 



of pure bad luck, the Japanese were given 
a sitting target. After 'a day of violent 
action, when all concerned tried to make 
up for what hindsight calls mistakes, the 
strength of the FEAF had been reduced 
by half. The way was open for the Japa- 
nese to begin landing operations.^^ 

The enemy had in fact made his first 
landing by the time of the main air at- 
tacks. A small force came ashore at dawn 
on Batan Island midway between Luzon 
and Formosa and immediately began 
work to set up an air base on an already 
existing strip to accommodate the rela- 
tively short-ranged Army fighters. The 
next day elements of two fighter regi- 
ments of the Japanese 5th Air Group 
were using the field and flying reconnais- 
sance and strike missions over northern 
Luzon, site of the next planned landings. 


On 9 December only a few enemy bomb- 
ers attacked, but these planes filtering 
through the early . morning darkness 
reached Nichols Field outside Manila un- 
scathed where their bombs increased the 
damage and added to the toll of Ameri- 
can planes. An all-out attack on the Ma- 
nila Bay area had been planned for the 

■* See Craven and Gate, op. cit., 201-213 and 
Fall of the Philippines, 79-90 for an examina- 
tion of the contradictory statements and chro- 
nology of misadventures that marked what may 
well have been the blackest day in the history 
of the Army Air Corps. 

^ Unless otherwise noted the material in this 
section is derived from Hart Narrative; 
USAFFE-USFIP Rept ; Moore Kept; RAdm 
F. W. Rockwell, Narrative of Naval Activities 
in the Luzon Area, lDec41-19Mar42, lAug42 
(located at NHD), hereinafter cited as Rock- 
well Narrative; Howard Rept; Jfth Mar Jnl; 
nth Army Rept; Fall of the Philippines; Rising 
Sun in the Pacific. 

second day of the war, but fog over 
Formosa prevented the take-off. Al- 
though the weather was again bad on the 
10th, the enemy naval squadrons were on 
their way to their targets by midmorn- 
ing.^" Radar and ground observers 
spotted the incoming flights, but to no 
avail. When the outnumbered American 
interceptors rose to greet the raiders, the 
enemy fighters swarmed all over them, not 
giving as good as they got, but more than 
making up for their losses whenever they 
downed one of the few remaining planes 
of FEAF. 

The Japanese bomber groups were head- 
ing for the best protected area in the Phil- 
ippines; almost all of the antiaircraft 
units in USAFFE were concentrated near 
Manila. But the gunners below had an 
insoluble problem; they had plenty of 
ammunition, but very little of it was fused 
so that it could reach above 24,000 feet. 
After a few false starts the enemy learned 
that they could bomb from heights of 
25,000 feet with relative impunity. There 
was a limited supply of mechanically- 
fused ammunition which could reach 
30,000 feet, but there were not enough 
such rounds to materially increase anti- 
aircraft defenses. 

The gunners of the 1st Separate Marine 
Battalion at Cavite scored a kill on 10 
December when they downed an over- 
eager dive bomber that strayed from the 
pack over Nichols Field, lured by the tar- 
get of two PBY's taking off from Sangley 
Point. But that was the end of it. The 
three-inch batteries turned back the first 
flight of bombers, which came in too low, 
but all subsequent flights approaching the 

^'HistSec, G-2, GHQ, FEC, Japanese Studies 
in WWII No. 13, NavOps in the Invasion of the 
Philippines, 15May46 (located at OCMH), 7. 



naval base stayed well out of range and 
the gunners were helpless spectators to the 
destruction that followed. ^^ 

Stick after stick of bombs rained down 
on the naval base as successive flights of 
bombers criss-crossed the area laying a per- 
ceptible pattern. Fires sprang up every- 
where as small dumps of ammunition and 
gasoline were hit and the old town of 
Cavite was soon a raging mass of flames. 
All the ships that could possibly get away 
from the yard headed out into the bay, but 
the bombers caught and fatally damaged 
a submarine and a mine sweeper. Every- 
one feared that the ammunition depot, 
which still had large quantities of powder 
and ammunition in it, would be hit, but 
the bombers missed their most promising 
target. Still, the fires being blown to- 
ward the depot from Cavite might touch 
it off,^^ and the rescue parties searching 
amid the flaming ruins for the hundreds of 
civilian casualties were in constant danger. 
Long after the raid was over, into the 
night and the early morning of the next 
day, the fires raged and Admiral Rockwell 
ordered all personnel to evacuate the base. 
Only a small group of men from Lieuten- 
ant Colonel Adams' battalion and a few 
Manila firemen remained. These volun- 
teers localized the fire and were able to 
save the commissary stores; the ammuni- 
tion depot soon was out of danger.^^ 

Admiral Hart had watched the air at- 
tack from the roof of his headquarters 
building in Manila and had seen the end 
of Cavite as a base of operations. Rock- 
well's damage report confirmed his obser- 

°' Hogaboom, 3. 

^LtCol J. V. Lyon Itr to CMC, 310ct56. 

"^ LtCol J. W. Keene, Narrative and tactical 
dispositions of the 1st SepMarBn at Cavite, 
8-26Dec41, n.d. 

vations. On the night of the 10th Hart 
ordered most of the remaining ships of the 
Asiatic Fleet still in Manila Bay to sail 
south to comparative safety. The next 
day he advised the captains of all mer- 
chant ships in the bay to get their vessels 
out while they still could; fortunately, 
only one merchantman out of 40 was 
caught by enemy bombers.^* The strong- 
est element of Hart's fleet, his 29 subma- 
rines, continued to operate from the bay 
for a short while, until Japanese control 
of the air made this base untenable. By 
the year's end only the submarine tender 
Ganopus and a small collection of yard 
craft, motor torpedo boats, and auxiliaries 
remained in Manila's waters. 

In the judgment of a naval historian of 
this period the Asiatic Fleet was "sadly 
inadequate" and therefore "unable to pre- 
vent the enemy from landing wherever he 
chose, or even to delay his efficient time- 
table of conquest." ^^ Nor were FEAF 
or the ground troops of USAFFE able to 
do the job. In some instances there was a 
temporary delay when planes hit the land- 
ing forces, but nowhere were the Japan- 
ese stopped and forced to turn back. On 
10 December two combat teams from the 
2d Formosa Regiment of the JfSth Division 
came ashore at Aparri in northern Luzon 
and at Vigan on the northwest coast. 
Their mission, which was to secure air- 
fields for use by Army planes, was success- 
ful. In a day the Japanese, despite the 
loss of several ships to American bombers, 
were firmly established ashore and in prac- 
tical control of the northern tip of Luzon. 
The one Philippine Army division in the 
area, the 11th, was responsible for the de- 

^ Hart Comments. 

' Rising Sun in the Pacific, 181. 



fense of the island north of Lingayen Gulf 
and was of necessity spread so thin that 
it could offer no effective resistance. 

The same situation held true in south- 
ern Luzon where the defending forces, two 
Philippine Army divisions, were com- 
pletely unable to cover all possible landing 
beaches. On 12 December, when a Japan- 
ese convoy carrying the advance assault 
detachment of the 16th Division, staged 
from the Palaus, reached Legaspi in south- 
eastern Luzon, there was nothing to op- 
pose their landing. The troops were 
ashore, had taken their airjReld objective, 
and were moving north by nightfall. In 
all there were less than 10,000 enemy troops 
ashore at this time, but they had behind 
them the rest of the Fourteenth Army and 
command of the sea and air to insure its 
arrival on schedule. 

The heavy air attacks of the 8th and 
10th were only harbingers of further 
aerial assaults. Reinforced by Army 
fighters and bombers operating from 
newly-seized airfields, the naval planes of 
the Formosa-based Eleventh Air Fleet 
spread out over Luzon seeking new targets. 
The first turn of Olongapo and the 4th 
Marines came on 12 December, the day 
that marked the end of effective U. S. air 

A flight of Japanese fighters followed 
the PBY's based at Olongapo into their 
anchorage after the flying boats had made 
a fruitless search for a supposed enemy 
carrier task force. The enemy pilots 
caught the seaplanes at their moorings and 
destroyed them all. As the Japanese 
strafed the naval station Marine machine 
gunners attempted to bring them down; 
Colonel Howard noticed that the tracers 
of Company H's .30's seemed to be "bounc- 
ing off these planes indicating sufficient 

armor plate to prevent penetration."^* 
The enemy attacked again on the 13th, 
this time bombing from altitudes beyond 
the range of the Marine automatic 
weapons. The few hits scored were all in 
the town of Olongapo ; there was no dam- 
age to the naval station and only a few 
Marine casualties. The Filipinos who ig- 
nored the air raid warning suffered heav- 
ily; a bomb hit right in the midst of a 
large group of townspeople who were 
"standing under a tree watching the per- 
formance," " killing 22 and wounding at 
least as many more. Although alarms 
were frequent thereafter, the Japanese did 
not attack again until the 19th and then 
their aim was bad and they liberally 
plastered the bay with bombs. 

During this period, while the original 
Japanese landing forces were advancing 
toward Manila, top-level discussions were 
held between Hart and MacArthur and 
their staffs regarding employment of the 
4th Marines.^^ On 2'0 December, Mac- 
Arthur formally requested that the regi- 
ment be assigned to his command "as de- 
velopments of the Navy plan can make it 
available."^ Admiral Hart concurred 
and directed Howard to report to 
USAFFE for such employment as Mac- 

^ Howard Rept, 9. 

" Jackson. 

''Although the war plan for the Philippines 
had long called for the available Marines to be 
assigned to the defending ground forces under 
Army control, USAFFE made no effort in the 
first few days of the war to contact CinCAF 
regarding his Marine forces. After a reminder 
from Hart, "there came back a request to send 
one battalion into Manila City to guard USAFFE 
Headquarters. Feeling that it would be a 
wrong use of the best infantry available, [Hart 
issued] no order to that effect." Hart Comments. 

'" Copy of CG, USAFFE Itr to CinCAF, 20Dec41, 
in 4th Mar Jnl, 237 ; Clement Rept, 6. 



Arthur might deem necessary in the de- 
fense of Luzon.*° In a covering memo to 
Rockwell, he pointed out that the assign- 
ment of the 4th Marines was the sole 
commitment that he had made and that 
he had verbally made it clear that it was 
his policy that excess naval personnel be 
organized and equipped and then "fed up 
into the combat areas on shore with the 
Fourth Regiment of Marines. A com- 
mand exercised over them by the Army 
would normally be via C. O. Fourth 
Marines." *^ 

The Navy Department had directed 
Hart on his departure from the Philip- 
pines to place all naval personnel, muni- 
tions, and equipment at the disposal of 
USAFFE. Rockwell, who was to relieve 
Hart as senior naval officer in the Philip- 
pines, nominally retained independent 
status. He adhered firmly, how^ever, to 
the principle of unity of command and 
cooperated closely with MacArthur's 

On 22 December, the reinforced Japa- 
nese j!i.8th Division landed in Lingayen 
Gulf. It was the logical landing point 
for any force whose objective was Manila, 
for the gulf stood at the head of a broad 
valley leading directly to the capital. The 
landing was expected and it was resisted, 
but the combined efforts of American air, 
submarine, and ground forces could not 
prevent the Japanese from putting their 

"According to the Hart Narrative, 45, one 
plan considered for the employment of the 4th 
Mar entailed the brigading of the regiment (re- 
inforced by a naval battalion from Mariveles) 
and a constabulary regiment under a Marine 
brigade commander. The Marines were to fur- 
nish oflBcers and NCO's to the Constabulary unit. 
Time did not permit the execution of this scheme. 

"CinCAF memo to ComSixteen, 22Dec41, in 
Wi- Mar Jnl, 239. 

men ashore and effecting a juncture with 
Vigan-Aparri landing forces which had 
driven down from the north. Resistance, 
although sparked by the 26th Cavalry of 
the Philippine Scouts, was spotty and in- 
effectual, and the Japanese soon proved 
that the partially-trained men of the 
Philippine Army were not yet a match 
for their troops. Covered by the Scout 
'cavalrymen, the Filipino reservists fell 
back in disorder to reorganize in positions 
below the Agno River. 

The enemy was ready to drive on Manila 
from the north. 

On the 24th, the last major assault ele- 
ment of the Fourteenth Army,, the 16th 
Division from the Ryukyus, landed at 
Lamon Bay only 60 miles cross-island 
from Manila. Here the story was much 
the same as Lingayen Gulf. The enemy 
overwhelmed scattered, ill-trained troops 
and made good his beachhead. The Amer- 
ican South Luzon Force began to fight 
a delaying action along the roads lead- 
ing to Manila. The decision, however, 
had already been made to declare the capi- 
tal an "open city," and the troops were 
headed for the Bataan Peninsula. 

Bataan, northern arm of Manila Bay, 
had long been considered the ultimate 
stronghold in a defense of Luzon. While 
it was held, and with it the fortified islands 
across the mouth of the bay, no enemy 
could use the harbor, and it was possible 
to gain succor from friendly naval forces 
which might break through a blockade. 
Mac Arthur had rejected the concept of a 
static, last-ditch defense when he took 
over USAFFE and had expected with the 
forces underway to him from the States 
and trained Philippine Army divisions to 
be able to repulse or contain enemy land- 
ing attempts. When the Japanese won 



control of the sea and air, however, he lost 
all chance for successful execution of his 
orders to "attack and destroy" *^ any land- 
ing force, and he was forced to adopt the 
only course of action that would save his 
army : a desperation withdrawal to Ba- 
taan. He made the fateful decision on 23 
December and the following day prepara- 
tions to effect it were begun. 

Basically the withdrawal plan called for 
Major General Jonathan C. Wainwright's 
North Luzon Force to fight a series of 
delaying actions in the central island plain 
which would allow the South Luzon Force 
(Major General George M. Parker, Jr.) 
to reach the peninsula. Then Wain- 
wright's units would pull back to Bataan, 
join forces with Parker, and stand off the 
enemy. In the time gained by the dejay- 
ing actions, USAFFE would make every 
effort to augment the supplies already 
gathered in scattered dumps on Bataan 
with food, ammunition, weapons, and 
equipment from installations in the 
Manila area. 

The role of the 4th Marines in this plan 
was laid out for Colonel Howard in a 
series of conferences which took place in 
Manila on 24 December. Admiral Hart, 
who was preparing to leave for Java the 
following day, informed the Marine com- 
mander that the 1st Separate Battalion 
would be added to his regiment as soon 
as it could clear Cavite and that he was 
to report immediately to MacArthur for 
duty. At USAFFE headquarters, amidst 
the bustle attendant on its move to Cor- 
regidor, Howard got a cordial welcome 
from his new chief and then received 
orders to move the 4th to Corregidor and 
take over its beach defenses. In a meeting 
with Admiral Rockwell, after he had 

" Wain icrigh t's Story, 28. 

made a final call on Hart, Howard was told 
to destroy the Olongapo Naval Station 
when he jjulled out. 

Mariveles at the southern tip of Bataan 
had been designated the assembly area and 
transshipment point for the Marine units 
and their supplies. The 1st Battalion had 
already spent two weeks in bivouac near 
the base weathering a series of air attacks 
and furnishing guard details, unloading 
parties, and dump construction crews. 
Two men were killed and three wounded 
on 24 December during a bombing raid 
that struck shipping in the harbor. It 
was an inauspicious portent for the recep- 
tion of the forward echelon of the regiment 
which left for Mariveles at 2200 that night. 

Shortly after the truck convoy had 
cleared Olongapo, Colonel Howard re- 
ceived warning from naval headquarters 
of an impending Japanese landing, and 
"sounds of motors could be distinctly 
heard from seaward" *^ in Subic Bay. All 
available men manned beach defense posi- 
tions, but fortunately the report proved 
false and the motors turned out to be those 
of American torpedo boats. Early on 
Christmas morning a message from Rock- 
well's new headquarters on Corregidor 
ordered Howard to expedite evacuation 
and destruction lest the regiment be cut 
off by advancing Japanese troops. The 
Philippine Army's 31st Division had 
pulled back to Bataan from its coastal 
positions northwest of Olongapo on the 
24th and the Marines' north flank was now 
open; a threat also existed to seaward, 
since the Army's coast defense troops were 
withdrawing from Fort Wint in Subic 
Bay. Motorcycle patrols ranging north 
of the base could find no sign of the enemy, 
however, and the movement of men and 

" Howard Rept, 11. 



supplies was completed without undue 
haste. At 1410 Howard's new CP opened 
outside Mariveles, and the fate of 
Olongapo was left in the hands of a demo- 
lition detail under Captain (later Major) 
Francis H. Williams. 

Using charges improvised from 300- 
pound mines, "Williams set out with his 
demolition gang to do a good job of eras- 
ing the Naval Station from the face of the 
globe." « They sank the hull of the old 
armored cruiser Rochester in the bay and 
blew up or burned everything of value 
except the barracks which closely bordered 
the native town.*^ The last supplies were 
loaded early Christmas evening, and the 
rear echelon pulled out with darkness. 

Christmas also saw the completion of 
destruction at Cavite where a Marine dem- 
olition party from the 1st Separate Bat- 
talion blew up or fired all remaining 
ammunition stocks and destroyed the sub- 
marine damaged in the 10 December air 
raid. The naval radio station near Sang- 
ley Point was already a shambles, for in 
a raid on 19 December enemy bombers 
leveled the buildings and set afire large 
quantities of gas and oil scattered in 
dumps throughout the surrounding area.*® 
Lieutenant Colonel Adams received orders 
on 20 December to evacuate the Cavite 
area, and for the next few days men and 
supplies were trucked to Mariveles; the 
Christmas day demolition detail was the 
last element to leave.*^ 

After darkness fell on 26 December, the 
first Marines to move to Corregidor, 14 
officers and 397 men of Adams' battalion, 

" Ferguson, 3. 

"The town's natives later burned their own 
houses and the barracks was consumed in the 
resulting Are. 

** Hogaboom, 4. 

" Keene, op. cit., 10. 

made the seven and a half mile voyage 
from Mariveles' docks to North Dock on 
"The Rock." 


The four islands that guarded the mouth 
of Manila Bay were fortified in the decade 
prior to World War I before air power 
changed the concept of coastal defense. 
Most of the powerful 14- and 12-inch guns 
were sited in open emplacements for the 
purpose of repelling an invasion from the 
sea. Disarmament treaty obligations and 
drastically reduced defense expenditures 
in the period between the wars allowed 
little concession to be made to the threat 
of air attack. Some antiaircraft guns were 
added to the fort's defenses, however, and 
a start was made toward providing under- 
ground bombproof shelters, especially on 
Corregidor (Fort Mills). 

Corregidor was at its closest point just 
a little over two miles from the tip of 
Bataan Peninsula. The island was tad- 
pole-shaped, three and a half miles long 
and one and a half miles wide at its head. 
This wide area, called Topside, loomed 
high above the rest of the island, its 500- 
foot cliffs dropping sharply to a narrow 
beachline. Most of the coast defense bat- 
teries and permanent quarters were located 
here, and the only access routes to the top 
from the western shore were two ravines, 
James and Cheney. East of Topside, 
along the neck of land that connected the 
tadpole's head and tail, was Middleside, a 
plateau which held several more battery 
positions and permanent buildings. A 

*' Unless otherwise noted the material in this 
section is derived from Moore Rept; Howard 
Rept; 4th Mar Jnl; Philippine AirOpsReo; Fall 
of the Philippines. 



third ravine, Ramsay, which led from the 
southern beaches to Middleside, was a crit- 
ical defensive point. (See Map 8, Map 

A logically-named third distinctive por- 
tion of the island, Bottomside, consisted of 
the low ground occupied by the industrial 
and dock area and the native town of San 
Jose. The nerve center of Luzon's de- 
fenses was an extensively-tunnelled hill, 
Malinta Hill, which rose directly east of 
San Jose. The headquarters of Mac- 
Arthur, Rockwell, and Major General 
George F. Moore, commanding the forti- 
fied islands, were all eventually located in 
the tunnels and laterals that spread out 
beneath the hill. From Malinta the long, 
low, narrow tail of the island bent away 
to the east; a light plane landing strip, 
Kindley Field, had been built along the 
spine of the tail. 

The other three island forts, Hughes, 
Drum, and Frank, complemented the de- 
fenses of Fort Mills, and Marines served 
as part of the beach defense troops on all 
but the last named. Caballo Island (Fort 
Hughes), a quarter-mile square in area, 
stood less than two miles from Corregi- 
dor; its low-lying eastern shore rose 
abruptly to a 380-foot height which con- 
tained most of the battery positions. Four 
miles south of Caballo was the "concrete 
battleship," Fort Drum. Tiny El Fraile 
Island had been razed to water level and on 
its foundation a steel-reinforced concrete 
fortress had been erected with sides 25- 
to 36-feet thick, and a top deck 20-feet 
deep. Two case-hardened steel gun tur- 
rets, each sporting a pair of 14-inch guns, 
were mounted on the deck and the sides 
of the fort boasted four 6-inch gun case- 
ments. Its garrison could be completely 
contained within its walls, and "the fort 

was considered impregnable to enemy at- 
tack."*® The island which seemed most 
vulnerable to assault was Carabao (Fort 
Frank) which lay only 500 yards from the 
shore of Cavite Province. However, since 
some of its guns were capable of firing in- 
land and most of its shoreline was ringed 
with precipitous cliffs, the job of taking 
Fort Frank promised to be quite a task. 
General Moore later ndted that "the fort- 
resses were not designed to withstand a 
landing attack from adjacent shores sup- 
ported by overwhelming artillery em- 
placed thereon ;" ^^ and that of his big guns 
only the turrfets of Fort Drum, the 12-inch 
mortars, -and two 12- inch long-range guns 
were capable of all-round fire. A tabula- 
tion of the major coast defense armament 
of the forts shows : 


Number of Guns 





14" euns. 














12" guns 

12" mortars 

10" guns 

8" guns 

6" guns. . 


155mm guns 

3" guns 


The forts had in addition a small number 
of 75mm beach defense guns. For anti- 
aircraft defense, including tied-in bat- 
teries on southern Bataan, there were 17 
searchlights, 40 3-inch guns, and 48 .50 
caliber machine guns." 

Marines from the 1st Separate Bat- 
talion were able to add a few .50 caliber 

' Moore Rept, 4. 
' Ibid., 9. 
Ihid., Annex C 



machine guns and a battery of four 3-inch 
guns taken from Cavite to the antiaircraft 
defenses, but the primary function of the 
battalion was now that of infantry. It 
was reorganized and re-equipped at Mari- 
veles to fill the role of the missing battalion 
of the 4th Marines ; the formal change of 
title to 3d Battalion, 4th Marines came 
on 1 January. 

With the exception of Batteries A and 
C and the radar detachment of Adams' 
battalion which remained on Bataan, the 
whole of the 4th Marines moved to Cor- 
regidor in successive echelons on the nights 
of 27 and 28 December. Enough rations 
for 2,000 men for six months, ten units 
of fire for all weapons, two years supply 
of summer khaki, and the medicines and 
equipment to outfit a 100-bed hospital ac- 
companied the move. Fortunately, the 
Quartermaster, Major Ridgely, dispersed 
these supplies in small, scattered dumps as 
they arrived and they emerged relatively 
unscathed from the first Japanese air raid 
on Corregidor. 

Many of the Marines in the bamboo 
jungles surrounding Mariveles, who had 
to shift camp constantly to avoid bombing 
and sleep "on the ground near a foxhole or 
some convenient ditch into which [they] 
could roll in the event of an air attack" ^^ 
looked forward to moving to The Rock. 
They had "watched the Jap bombers steer 
clear of its antiaircraft barrages" and it 
had been pointed out to them "that Cor- 
regidor 's antiaircraft was so good that the 
Japs had not even dared to bomb it — 
yet !" ^^ An additional lure of the island 

" Statement of Lt(jg) R. G. Hetherneck in Cdr 
T. H. Hayes Rept on MedTactics, 4th Regt USMC, 
7Dec41-6May42, 15Feb46, 79, hereinafter cited as 
Hayes Rept. 

"LtCol R. F. Jenkins, Jr., Personal Experi- 
ences 28Dec41-6May42, n.d., 1, hereinafter cited 
as Jenkins. 

to some men was the vision of a Gibraltar, 
and they talked knowingly of the (non- 
existent) intricate underground system of 

At 0800 29 December, Colonel Howard 
reported to General Moore for orders as 
Fort Mills' beach defense commander and 
then started out to make a reconnaissance 
of the island. His men, temporarily 
quartered in Middleside Barracks, were 
startled to hear the air raid sirens sound 
shortly before noon. No one paid too 
much attention to them as Corregidor had 
never been bombed, but soon their trust- 
ing attitude changed. "All hell broke 
loose," and as one 1st Battalion officer 
described the scene, "there we were — the 
whole regiment flat on our bellies on the 
lower deck of Middleside Barracks." ^^ 

The Japanese planes, 40 bombers of the 
5th Air Group with 19 covering fighters, 
attacked at 1154. For the next hour a 
parade of Army aircraft flew the long 
axis of Corregidor dropping 200- and 500- 
pound bombs from 18,000 feet, and dive 
bombers attacked the antiaircraft bat- 
teries, strafing as they plunged down. At 
1300, the Army planes gave way to the 
Navy and bombers of the Eleventh Air 
Fleet continued to attack until 1415. None 
of FEAF's few remaining fighters, which 
were being saved for vital reconnaissance 
missions, took to the air, but Corregidor's 
gunners exacted a good price from the 
enemy — 13 medium bombers fell to the 
3-inchers and the .50 calibers shot down 
four of the dive bombers in a vivid demon- 
stration of the folly of flying within reach 
of these guns. But the damage done by 
the enemy was considerable. 




Almost all of the barracks and head- 
quarters buildings and a good half of the 
wooden structures on the island were bat- 
tered, set afire, or destroyed. A thick pall 
of black smoke and clouds of dust obscured 
the island from observers on Bataan, and 
the detail left to load out Marine supplies 
wondered at the fate of the regiment in 
the center of this maelstrom. ^® The 
casualty score was miraculously low, only 
one man killed and four wounded. With a 
single exception, bombs used by the Japa- 
nese did not penetrate all the way through 
to the bottom deck of the concrete bar- 
racks, but the building, shaken repeatedly 
by hits and near misses, was a shambles. 
In all, the island's defenders suffered about 
100 casualties, and 29 December marked 
the end of "normal" above ground living 
for The Rock's garrison. 

As soon as the air raid was over, Howard 
assigned beach defense sectors to his bat- 
talions, and the troops moved out to their 
new bivouac areas before dark. The 1st 
Battalion, 20 officers and 367 enlisted men 
under Lieutenant Colonel Curtis T. 
Beecher, drew a possible enemy landing 
point — the East Sector which included 
Malinta Hill and the island's tail. The 
beaches of Bottomside and most of Mid- 
dleside (Middle Sector) up to a line in- 
cluding Morrison Hill and Ramsay Ra- 
vine were occupied by Lieutenant Colonel 
Adams' battalion with 20 officers and 490 
men. The defense of the rest of the shore- 
line of Corregidor was the responsibility 
of the 2d Battalion (Lieutenant Colonel 
Herman R. Anderson) which mustered 18 
officers and 324 enlisted men. A general 
reserve of 8 officers and 183 men, formed 
from the Headquarters and Service Com- 

panies, commanded by Major Stuart W. 
King," bivouacked in Government Ravine, 
on the southern shore of the island below 
Geary and Crockett Batteries. 

Not all of Adams' battalion was as- 
signed to the Middle Sector; besides the 
units left on Bataan, the 3d Battalion 
furnished most of the other special de- 
tachments. One platoon (1 officer and 28 
men) with four .50 caliber machine guns 
and a second (1 officer and 46 men) with 
four .30's left for Fort Hughes on the 
30th to bolster the antiaircraft and beach 
defenses ; the 2'd Battalion added ten men 
and four more .30 caliber machine guns 
to the beach defenses on 3 January.'^* Fort 
Drum got a section of 15 men and two .50's 
to augment its crew. A third antiaircraft 
platoon with six .50's (1 officer and 35 
men) was directly assigned to Fort Mills' 
air defenses and attached to a similarly- 
equipped battery of the 60th Coast Artil- 
lery which was emplaced near the Topside 
parade ground. 

By 1 January the pattern had been set 
for the Marines' duties on Corregidor. 
The men were digging in, stringing barbed 
wire, emplacing their 37mm's, mortars, 
and machine guns, and tying-in for a co- 
ordinated and protracted defense. Ahead 
lay more than four months of waiting and 
preparation for a battle, months in which 
more than one survivor likened life on 
Corregidor to existence in the center of a 
bull's eye. 

' Fergnson, 6. 

'''' Hotvard Kept, 14 lists Maj (then Capt) Max 
W. Schaeffer as commander of the regimental 
reserve at this point ; however, the contemporary 
muster rolls at HQMC and survivors' comments 
indicate that Maj King held this position until 
17Feb42 when Maj Schaeffer took over and King 
became ExO of the beach defense force at Fort 

^ Lyon Itr, op. cit. 

Bataan Prelude 



The Japanese did not confine their oper- 
ations in the Philippines to Luzon, but 
December landings on Mindanao and Jolo 
Islands were made primarily to secure 
bases for further attacks against Borneo. 
Here again the hastily mobilized Philip- 
pine Army reservists and Constabulary 
troopers were no match for the assault 
forces, and the enemy made good his lodg- 
ment. Offensive operations in the south, 
however, were limited by the fact that 
General Homma did not have sufficient 
troops to press a campaign on two fronts. 
The main strength of the Fourteenth 
Army remained on Luzon to win control 
of Manila Bay. 

The original Japanese operation plan 
for Luzon had contemplated its occupa- 
tion by the end of January and had pro- 
vided for a mop-up force of one division 
and one brigade with a small air support 
unit. Shortly after Homma's troops en- 
tered Manila on 2 January, he received 
orders to expedite the withdrawal of the 
4^th Division and the 5th Air Group 
which were needed to reinforce stepped-up 
operations against Java and on the Asian 
mainland. In short return for these troop 
losses Homma got the 65th Br^igade from 
Formosa, originally assigned to mop-up 
and police duties. The brigade, which 

' Unless otherwise noted the material in this 
section is derived from VSAFFE-USFIP Kept; 
Vfth Army Kept; Fall of the Philippines. 

landed on Lingayen Gulf beaches on 1 
January, was selected by Homma as his 
Bataan assault force and reinforced with 
an infantry regiment from the 16th Di- 
vision and tank, artillery, and service 
troops from army reserves. (See Map 7, 
Map Section) 

The necessary reorganization of the 
65th Brigade for combat and its movement 
into jump-off positions gave MacArthur 
time to establish an initial defense line. 
On 7 January he reorganized his forces 
into two corps and a rear area service 
command. Wainwright was given I Phil- 
ippine Corps with responsibility for hold- 
ing the western front, and Parker became 
II Philippine Corps commander with his 
troops manning defenses on the Manila 
Bay side of Bataan. More than 80,000 
men were now bottled up on the peninsula 
and some 50,000 held positions on or near 
the initial defense line. These were im- 
pressive figures, and in paper strength 
the Bataan defenders outnumbered the 
Fourteenth Arrwy which had about 50,000 
troops under its command. 

Military superiority depends, however, 
on many other factors besides relative 
troop strength. The conglomerate Amer- 
ican-Filipino forces, completely cut off 
from effective relief, had limited supplies 
of rations, medicines, weapons, ammuni- 
tion, and equipment. By contrast, the 
enemy's control of the sea and air gave 
the Japanese an unmatchable resupply 
and reinforcement potential. Even when 




the Fourteenth Army''s fortunes were at 
their lowest ebb, the enemy troops could 
reasonably expect rescue and relief. 

On 5 January MacArthur, in a move to 
conserve dwindling food stocks, had cut all 
troops on Bataan and the fortified islands 
to half rations.^ This order was un- 
doubtedly the most significant given in 
the campaign. It prolonged the fighting 
for weeks, until Bataan's defenses eventu- 
ally collapsed. Men sapped by malnu- 
trition and its attendant diseases, for 
which there were no medicines, could 
resist no more. 

In launching their initial attack on Ba- 
taan the Japanese did not expect that the 
reinforced 6Sth Brigade would have much 
trouble defeating the American-Filipino 
forces. The enemy was flushed with his 
successes and "completely ignorant con- 
cerning the terrain of Bataan Peninsula." ^ 
Homma's intelligence officers had under- 
estimated MacArthur's strength by half, 
had given their commander a distorted 
picture of Filipino morale, and had formu- 
lated an altogether incorrect estimate of 
the defensive situation on Bataan. The 
Fourteenth Army staff had : 

. . . optimistically presumed that, considering 
its position relative to Corregidor Island, the 
enemy would offer serious resistance at the 
southern end of the Peninsula with Mariveles as 
a nucleus, withdrawing later to Corregidor 
Island. Taking this for granted, the threat of 
enemy resistance was taken lightly.' 

Bataan Peninsula was an ideal position 
from the viewpoint of a force committed 
to a last-ditch stand. Thirty miles long at 

^ "Actually, the troops on Bataan received 
about one-third ration." V8AFFE-USFIP Rept, 

" 14th Army Rept, 90. 

* Ibid., 91. 

its deepest point and 25 miles wide at its 
base, the peninsula tapered to an average 
width of 15 miles. Numerous streams, 
ravines, and gullies cut up the interior 
and thick jungle growth blanketed every- 
thing. A spine of mountains running 
northwest to southeast split Bataan 
roughly in half. The dominant features 
in the north were Mt. Natib (4,222 feet) 
and its companion Mt. Silanganan (3,620 
feet), and in the south, Mt. Bataan (4,700 
feet),, which commanded the Mariveles 
area. Although numerous trails criss- 
crossed the peninsula, only two motor 
roads existed, one running along the coast 
and the other over t\\^ saddle between the 
mountain masses. The western coast line 
was uneven with many promontories 
formed by mountain ridges; the eastern 
coast was more regular and open but be- 
came hilly and rugged in the south. (See 
Map 7, Map Section) 

The final defense line selected by 
USAFFE was midway down the penin- 
sula, anchored on the towns of Bagac and 
Orion, and generally along the trace of 
the cross-peninsula motor road. It was 
the necessity of covering the preparation 
of this area for defense and the need to use 
the road as a supply route as long as pos- 
sible that dictated the occupation of the 
initial defense line. Stretching across the 
peninsula just above the point where it 
narrowed, this position had a grave nat- 
ural weakness. The corps boundary ran 
along the Natib-Silanganan mountain 
mass which pierced the defenses and pre- 
vented liaison or even contact between 
Wainwright's and Parker's men. The 
Japanese attempt to crack this defense 
line eventually involved landings far be- 
hind the front and brought the Marines 
at Mariveles into action. 



SYMBOLIC OF JAPANESE SUCCESSES in the early stages of the war is this photograph taken on Mt. 
Limay on Bataan during the fighting in the Philippines. ( SC 334265 ) 




Although the farthest distance from the 
rear boundaries of the corps areas to the 
southern shore of Bataan was only ten 
miles, the defensive problem facing Brig- 
adier General Allan C. McBride's Service 
Command was acute. With a relatively 
few men McBride had to guard over 40 
miles of rough, jungle-covered coast line 
against enemy attack. A successful am- 
phibious thrust which cut the vital coastal 
supply road could mean the prompt end 
of the battle for Bataan. To protect the 
east coast he had the newly-organized 2d 
(Constabulary) Division; on the west 
coast he had a motley composite force of 
service troops and planeless pursuit squad- 
rons converted to infantry, backed up by 
a few elements of the 7lst Division and 
a Constabulary regiment. Responsibility 
for the security of the naval reservation 
at Mariveles remained with the Navy. 

In order to provide protection for 
Mariveles and support the Army in the 
defense of the west coast, Admiral Rock- 
well on 9 January directed Captain John 
H. S. DesSez, commander of the section 
base, to form a naval battalion for ground 
combat. The senior naval aviator remain- 
ing in the Philippines, Commander Fran- 
cis J. Bridget, was appointed battalion 
commander and he immediately set about 
organizing his force. For troops he had 
about 480 bluejackets including 150 of his 
own men from Air, Asiatic Fleet, 130 

^ Unless otherwise noted the material in this 
section is derived from Rockwell Narrative ; 16th 
NavDist War Diary, 8Dec41-19Feb42 (located 
at NHD) ; Clement Rept; Cdr F. J. Bridget Kept 
to ComSixteen, Action at Longoskawayan Point, 
9Feb42 (located at NHD), hereinafter cited as 
Bridget Rept; Capt E. L. Sackett, USN, "History 
of the USS Canopus," 28Apr47 (located at NHD), 
hereinafter cited as Canopus Hist; Hogaboom. 

crewmen from the submarine tender 
Canopus, 80 sailors from the Cavite Naval 
Ammunition Depot, and 120 general 
duty men from Cavite and Mariveles. 
He was also assigned approximately 120 
Marines, members of Batteries A and C 
which had remained behind on Bataan 
under naval control when the rest of the 
1st Separate Battalion (now 3/4) had 
moved to Corregidor. 

The men of First Lieutenant William F. 
Hogaboom's Battery A had originally 
been slated to provide replacement and re- 
lief gun crews for Battery C (First Lieu- 
tenant Willard C. Holdredge) whose 3- 
inch guns were set up in a rice paddy 
between the town of Mariveles and the 
section base. But on 5 January Hoga- 
boom had received instructions from a 
USAFFE staff officer, "approved by naval 
authorities on the 'Rock','' ® to move his 
unit to the site of MacArthur's advance 
CP on Bataan where the Marines were 
to furnish the interior guard. This as- 
signment was short-lived, however, since 
Commander Bridget needed the men to 
serve as tactical instructors and cadres for 
the naval battalion, and on 14 January he 
directed Hogaboom to report back to 
Mariveles. To replace Battery A, 
ITSAFFE detached two officers and 47 
men from the 4th Marines ^ and sent 
them from Corregidor to Bataan where 
they guarded the advance headquarters 
until the end of the campaign. 

The most serious problem Bridget faced 
in forming his battalion was the lack of 
ground combat training of his blue- 
jackets. As the commander of the Can- 
opus, naturally an interested spectator, 
noted : 

° Hogaboom, 5. 
' 4th Mar Jnl, 2Q8. 



. . . perhaps two-thirds of the sailors knew 
which end of the rifle should be presented to 
the enemy, and had even practiced on a target 
range, but field training was practically a closed 
book to them. The experienced Marines were 
spread thinly throughout each company in hope 
that through precept and example, their qualities 
would be assimilated by the rest." 

Even after the formation of the naval 
battalion, the primary responsibility for 
antiaircraft defense of Mariveles still 
rested with the Marine batteries and only 
a relatively few men, mostly NCO's, could 
be spared to help train the bluejacket 
companies. Holdredge's 3-inchers re- 
quired at least skeleton crews and Hoga- 
boom's unit, after its return from 
USAFFE control, was directed to mount 
and man nine .50 caliber machine-gun 
posts in the hills around the harbor. 
Therefore, in both batteries the majority 
of men available for ground combat were 
sailors; Battery A joined one officer and 
65 bluejackets on 16-17 January and a 
Navy officer and 40 men joined Hol- 
dredge's battery on the 18th and 19th.* 
Throughout the naval battalion, training 
was confined to fundamentals as Bridget 
strove to qualify his men as infantry. As 
was the case so often in the Philippines, 
the time for testing the combat readiness 
of the jury-rigged battalion came all too 


In opening his atack on Bataan, General 
Homma committed the main strength of 
the 65th Brigade along the front of 
Parker's II Corps, figuring that the more 
open terrain along the east coast gave him 
a greater opportunity to exploit successes. 
By 11 January the Japanese had devel- 

oped and fixed Parker's defenses and were 
probing for weak spots preparatory to an 
all-out assault. It was inevitable that 
they found the open and highly vulnerable 
left flank. By 22 January Parker's posi- 
tion along the slopes of Mt. Natib had been 
turned and all reserves with the exception 
of one regiment had been committed to 
contain the penetration. In order to pre- 
vent the defending forces from being cut 
off, USAFFE ordered a general with- 
drawal to the Bagac-Orion defense line, 
to be completed by the 26th. 

The enemy advancing along the moun- 
tainous west coast did not contact Gen- 
eral Wainwright's forward positions until 
15 January. By that date, Homma, im- 
pressed by the lack of resistance in this 
sector, had already ordered the Wth In- 
fantry of the 16tK Division to reinforce 
and exploit the drive, strike through to 
the Bagac road junction, and gain the rear 
of Parker's corps. Although I Corps had 
been stripped of reserves to back up the 
sagging eastern defense line, Wainwright's 
front-line troops were able to stand off the 
initial Japanese assaults. When Homma's 
fresh troops attacked on the 21st, however, 
they effected a lodgment behind the front 
which eventually made withdrawal toward 
Bagac mandatory. The local Japanese 
commander, encouraged by his success, de- 

• Canopus Hist, 14. 

' Col W. F. Prickett Itr to CMC, October 1956. 

" Unless otherwise noted the material in this 
section is taken from V8AFFE-U8FIP Kept; 
nth Army Rept; 4th Mar Jnl; Clement Rept; 
Bridget Rept; Hogaboom; Btry A, US NavBn, 
Narrative of events, 2Feb42 ; 2dLt M. E. Peshek 
Itr to CO, 2/4, "Report of operation of Marine 
Detachment sent to Bataan on 25 January 1942," 
2Feb42; GunSgt H. M. Ferrell Itr to CO, 1/4, 
"Temporary Duty of Mortar Platoon, vicinity 
of Mariveles, Bataan, Philippine Islands, from 
.January 25, 1942, to January 30, 1942, Inclu- 
sive," 31 Jan42 ; Fall of the Philippines. 



cided on a shore-to-shore amphibious as- 
sault which would hit the Bataan coastal 
road about four miles below Bagac. 

Embarking after dark on the night of 
22-23 January, the enemy's 900-man land- 
ing, force {£d Battalion, 20th Infantry) 
started out for its objective. It never 
arrived. En route two launches of the 
battalion's boat group were discovered and 
sunk by an American torpedo boat " and 
it is possible that these attacks were in- 
strumental in scattering the remainder 
of the landing force. In any event, the 
enemy boats lost their bearings completely 
in the darkness. Instead of landing on 
the objective, two-thirds of the unit landed 
at Quinauan Point, eight miles south of 
Bagac. The remainder of the battalion, 
7 officers and 294 men, came ashore at 
Longoskawayan Point, a finger-like prom- 
ontory only 2,000 yards west of Mari- 
veles. (See Map 7, Map Section) 

The Longoskawayan landing force was 
not discovered immediately, and the enemy 
had time to advance along jungle-matted 
cliffs and reach Lapiay Point, the next 
promontory to the north. The Japanese 
patrols headed inland from Lapiay for 
Mt. Pucot, a 617-foot hill which com- 
manded both the west coast road and the 
landing site. The first word of the pres- 
ence of the enemy in his defense sector 
reached Commander Bridget at 0840 on 
23 January when the small lookout detach- 
ment he had posted on Pucot was driven 
from its position by enemy machine-gun 

Bridget immediately phoned Hogaboom 
and Holdredge, directing both officers to 
send out patrols. Hogaboom, closest to 

" ComMTBron-3. Kept of Act of USS PT-34 on 
the night of 22-23 Jan42, 27Feb42 (located at 

the scene of action, sent one bluejacket 
platoon under Lieutenant (junior grade) 
Leslie A. Pew directly to Pucot while he 
led a second platoon himself in a sweep 
through the ridges south of the hill. Pew's 
platoon deployed as it approached the hill- 
top, attacked through scattered rifle and 
machine-gun fire, and secured the high 
ground without difficulty. The Japanese 
offered only slight resistance and then 
faded out of contact. 

South of the hill Hogaboom ran into a 
platoon from Battery C which had had a 
brush with the Japanese and taken a couple 
of casualties, but again the enemy had 
disappeared. The story of light firing 
and no firm resistance was much the same 
from the rest of the probing patrols which 
Bridget ordered out on the 23d ^ the Japa- 
nese evidently were still feeling out the 
situation and were not as yet disposed to 
make a stand or an attack. At dusk the 
patrols assembled on the mountain and 
set up a defense line along its crest and the 
ridges to the south facing Lapiay and 
Longoskawayan Points. 

During the day Bridget had called on 
the Service Command for reinforcements 
but few men could be spared as most re- 
serves already had been committed to con- 
tain the larger landing force at Quinauan 
Point. For infantry he got the 3d Pur- 
suit Squadron and 60 men from the 301st 
Chemical Company whom he put into a 
holding line above Lapiay Point and on 
the north slope of Pucot ; for fire support 
lie received one 2.95-inch mountain pack 
howitzer and crew from the 71st Division. 
By nightfall the chemical company had 
tied in with the pursuit squadron on its 
right and with Battery A atop Pucot on 
its left ; platoons from Battery C, the Air, 
Asiatic Fleet Company, and the Naval 



Ammunition Depot Company held the rest 
of Pucot's crest and the southern ridges 
which blocked off the landing area from 
Mariveles. ^^ None of the naval battalion 
units was at full strength and none of the 
platoons strung out along the ridges was 
strictly a Navy or a Marine outfit. Sailors 
predominated but Marines were present all 
along the line, mostly as squad leaders and 
platoon sergeants. The composition of 
that part of the battalion which got into 
action became even more varied as the 
battle shaped up, and eventually about a 
third of the men on the front-line were 
drawn from the Marine batteries. 

Bridget's men got their first real taste 
of the blind fighting of jungle warfare on 
the 24th. Hogaboom led a patrol down 
the bluff above Lapiay Point and ran head- 
on into an enemy machine gun firing from 
heavy cover. Grenades thrown at the gun 
exploded harmlessly in a tangle of lush 
vegetation which screened it from view; 
the men were being fired upon and they 
were replying, but sound rather than sight 
was the key to targets. When reinforce- 
ments arrived later in the day an attempt 
was made to establish a holding line across 
the point, but the Japanese opened up with 
a second machine gun and steady rifle fire. 
Then they began dropping mortar and 
howitzer shells among Hogaboom's group. 

'^ An oflScer of 3/4 who knew many of the sur- 
vivors of this action, later wrote an article de- 
scribing the battle. He maintained that the Army 
detachments mentioned never joined, although 
they may have been assigned, and that the purely 
naval companies were not used as such. Instead 
the sailors who could be spared joined one of the 
Marine units. While this story is quite credible, 
in this instance the contemporary Bridget Rept 
has been used as a guide. See LtCol W. F. 
Prickett, "Naval Battalion at Mariveles," MC 
Gazette, June 1950, 40-43. 

The Marine officer ordered a withdrawal 
to the previous night's positions. 

The source of the mortar and howitzer 
fire was Longoskawayan Point where a 
patrol led by Lieutenant Holdredge had 
encountered the main body of Japanese. 
His two-man point had surprised an enemy 
group setting up a field piece in a clear- 
ing and opened up with a rifle and a BAR, 
dropping about a dozen men around the 
gun. The Japanese reaction was swift, 
agreeing with the BAR-man's evaluation 
that the surprise fire "ought to make them 
madder'n hell."" The patrol fell back, 
fighting a rear guard action until it cleared 
the area of the point's tip, and then it 
retired to the ridges. After the day's 
action Hogaboom and Holdredge com- 
pared notes and estimated that they faced 
at least 200 well -equipped enemy troops 
in strong positions ; they informed Bridget 
of their conclusion that it would take a 
fully-organized battalion with supporting 
weapons to dislodge them. 

On the morning of 25 January, 
USAFFE augmented Bridget's force by 
sending him a machine-gun platoon and 
an 81mm mortar platoon from the 4th 
Marines on The Rock. The two mortars 
immediately set up on a saddle northwest 
of Pucot and, with Hogaboom spotting for 
them, worked over the whole of Lapiay 
and Ix)ngoskawayan Points; direct hits 
were scored on the positions where the 
Japanese had been encountered the day 
before. A midaf ternoon patrol discovered 
that the enemy had evacuated Lapiay, but 
it was soon evident where they had gone. 
Holdredge led a combined force of several 
platoons against Longoskawayan Point 
and ran into a hornet's nest. He himself 
was among those wounded before the pla- 

' Quoted in Hogaboom, 10. 



toons could extricate thetnselves. Again 
the naval battalion occupied blocking posi- 
tions on the ridges east of the points for 
night defense. 

During the action of the 25th, USAFFE 
had changed the command structure in the 
rear service area and given the corps com- 
manders responsibility for beach defense 
throughout Bataan. In addition, Mac- 
Arthur had granted permission for the 
12-inch mortars on Corregidor to support 
Bridget's battalion. Shortly after mid- 
night, the giant mortars, spotted in by an 
observer on Mt. Pucot, laid several rounds 
on Longoskawayan Point. The daylight 
hours were spent in light patrol action 
while the battalion was readied for a full- 
scale attack on the 27th. General Wain- 
wright sent a battery of Philippine Scout 
75mm guns to support the drive. 

At 0700 on 27 January the mountain 
howitzer, the Marine mortars, the Scout 
75's, and Corregidor's 12-inch mortars 
fired a preparation on Longoskawayan, 
and a skirmish line of about 200 men, some 
60-75 of them-Marines, started to advance. 
The enemy reoccupied his positions as soon 
as the supporting fire lifted, and the 
jungle came alive with bullets and shell 
fragments. The right and center of the 
line made little progress in the face of 
heavy machine-gun fire. On tlie left 
where the going was a little easier a gap 
soon opened through which Japanese in- 
filtrating groups were able to reach the 
reserve's positions. 

During the resulting hectic fighting, the 
enemy opened up with mortar fire to 
herald a counterattack; fortunately, the 
4th Marines' 81's were able to silence this 
fire, but it was soon obvious that the naval 
battalion was in no shape to advance 
farther or even to hold its lines on Longo- 

skawayan. Bridget again authorized a 
withdrawal to the night defense lines on 
the eastern ridges. 

The solution to the problem of elimina- 
ting the Japanese beachhead arrived late 
that afternoon. Colonel Clement, who 
had come over from Corregidor to advise 
Bridget on the conduct of the Longo- 
skawayan action, had requested reinforce- 
ments from I Corps. Wainwright sent in 
the regular troops needed and the 2d Bat- 
talion of the 57th Philippine Scout Regi- 
ment relieved the naval battalion, which 
went into reserve. The 4th Marines' mor- 
tars and machine guns were assigned to 
the Scouts to support their operations. 
The oddly -assorted platoons of Bridget's 
battalion were not committed to action 
again, but they had done their job in con- 
taining the Japanese though outnumbered 
and outgunned. 

The Scouts spent 28 January in develop- 
ing the Longoskawayan position. On the 
29th they attacked in full strength with 
all the support they could muster. The 
mine sweeper Qvuil, risking an encounter 
with Japanese destroyers, came out from 
Mariveles and cruised offshore while Com- 
mander Bridget spotted for the 12-inch 
mortars and the 75mm guns. The ship 
closed from 2,200 to 1,300 yards firing 
point-blank at Japanese soldiers trying to 
hide out in the caves and undergrowth 
along the shores of the point." Ashore 
the Scouts, supported by machine-gun and 
mortar fire from the landing flanking the 
point, did the job expected of them and 
smashed through the enemy lines. By 
nightfall organized resistance had ended 
and the cost of taking Lapiay and Long- 

" CO USS Quail Rept of Act at Longoskawayan 
Pt morning of 29Jan42, 30Jan42 (located at 
NHD), L 



oskawayan had been counted. Bridget's 
unit had lost 11 killed and 26 wounded in 
action ; the Scout casualties were 11 dead 
and 27 wounded; and the Japanese had 
lost their entire landing force. 

During the next few days patrols, aided 
by ship's launches armored and manned 
by crewmen from the Cano^ms^ mopped 
up the area, killing stragglers and taking 
a few prisoners, but the threat to Mari- 
veles was ended. Similar action by ade- 
quately supported Scout units wiped out 
the Quinauan Point landing force by 7 
February. The major Japanese attempt 
to reinforce the beleaguered troops on 
Quinauan was beaten back by the com- 
bined fire of artillery, naval guns, and the 
strafing of the four P-40's remaining on 
Bataan. Elements of an enemy battalion 
which did get ashore on a point of land 
just above Quinauan on 27 January and 
2 February were also finished off by the 

By 13 February the last survivors of the 
amphibious attempts had been killed or 
captured. The make-shift beach defense 
forces which had initially contained the 
landings had barely managed to hold their 
own against the Japanese. They had had 
to overreach themselves to keep the enemy 
off balance and prevent a breakthrough 
while the troops of I and II Corps were 
falling back to the Bagac-Orion position. 
Once that line was occupied and Wain- 
wright could commit some of his best 
troops in sufficient numbers and with ade- 
quate support, the Japanese were finished. 
Discouraged by their amphibious fiasco, 
the enemy never again attempted to hit the 
coastal flanks of the American-Filipino 

At the same time the survivors of the 
landing attempts were being hunted down. 

the Japanese offensive sputtered to a halt 
in front of the Bagac-Orion line. The 
initial enemy advance on Bataan had not 
been made without cost, and the casualty 
rate now soared so high that the attack- 
ing troops were rendered ineffective. On 
13 February Homma found it necessary 
to break contact, pull back to a line of 
blocking positions, and to regroup his bat- 
tered forces. The lull in the Fourteenth 
Army''s attack was only temporary, how- 
ever, as Homma was promised replace- 
ments and reinforcements. When the 
second phase of the battle for Bataan 
opened, the scales were heavily tipped in 
favor of the Japanese. 

The detachment of Canofus crewmen, 
the sailors from the Cavite Naval Ammu- 
nition Depot, and the majority of the gen- 
eral duty men, nine officers and 327 en- 
listed men in all, were transferred to the 
4th Marines on Corregidor on 17-18 Feb- 
ruary. Commander Bridget and his naval 
aviation contingent moved to Fort Hughes 
on the 30th where Bridget became beach 
defense commander with Major Stuart W. 
King of the 4th Marines as his executive 
officer. Battery C of 3/4 remained at 
Mariveles to man its antiaircraft guns, 
but Battery A rejoined the regiment, with 
most of its men going to Headquarters 
Company to augment the regimental 

The assignment of the sailors of the 
naval battalion to Colonel Howard's com- 
mand accentuated the growing joint- 
service character of the Marine regiment. 
Small contingents of crewmen from 
damaged or sunken boats of the Inshore 
Patrol also had been joined and over 700 
Philippine Army air cadets and their offi- 
cers were now included in the 4th's ranks. 
These men, most of whom had never had 



any infantry training, were distributed 
throughout the companies on beach de- 
fense and in reserve where the experienced 
Marines could best train them by example 
and close individual instruction. No com- 
pany in the regiment retained an all- 
Marine complexion. 

The arrival of reinforcements on Cor- 
regidor and Caballo came at a time when 
the Japanese had stepped up their cam- 
paign against the fortified islands. On 6 
February, the first enemy shells, fired by 
105mm guns emplaced along the shore of 
Cavite Province, exploded amidst the 
American positions on all the islands. 
The reaction was swift and the forts re- 
plied with the guns that could bear. The 
counterbattery exchange continued 
throughout February and early March, 
occasionally waning as the Japanese were 
forced to shift to new firing positions by 
gunners on Forts Frank and Drum. The 
limited number of planes available to 
Homma made enemy bombers infrequent 
visitors during this period, and the Jap- 
anese concentrated on reducing the island 
defenses with artillery fire. In the first 
week of March, the American commander 
on Fort Frank received a demand for its 
surrender with a boast that the Cavite 
coast was lined with artillery and that : 

. . . Carabao will be reduced by our mighty 
artillery fire, likewise Drum ; after reduction 
of Carabao and Drum our invincible artillery 
will pound Corregidor into submission, batter 
it, weaken it, preparatory to a final assault by 
crack Japanese landing troops." 

The surrender note was unproductive 
for the enemy, but it was prophetic regard- 
ing the fate of Corregidor. 

Until the Japanese were ready to renew 
their assault on Bataan in late March, the 

severity of the enemy shellings from Cavite 
was not great enough to be effective in 
halting the construction and improvement 
of beach defenses on Corregidor. 
Trenches and gun positions lined the 
shores of Bottomside and the ravines lead- 
ing to Topside and Middleside from the 
beaches. Barbed wire entanglements and 
mine fields improvised from aerial bombs 
were laid across all possible approaches. 
The ordnance stores of the island were 
searched to provide increased firepower 
for the 4th Marines,^^ and guns were sited 
to insure that any landing force would be 
caught in a murderous crossfire if it at- 
tempted to reach shore. 

The thoroughness of the regiment's 
preparations was indicative of its high 
state of morale. The men manning the 
beach defenses, and to a lesser extent their 
comrades in the jungles of Bataan, never 
completely abandoned hope of rescue and 
relief until the very last days of their 
ordeal." Even when General MacArthur 
was ordered to leave the Philippines to 
take over a new Allied command in the 
Southwest Pacific, many men thought that 
he would return, leading a strong relief 
force. The senior commanders in the 
Philippines and the Allied leaders knew 
the truth, however, and realized that bar- 
ring a miracle, Luzon was doomed to fall. 
Only a few key men could be taken out of 
the trap by submarine, torpedo boat, or 

' Quoted in U8AFFE-U8FIP Kept, 40-41. 

" One source of beach defense guns was the 
sub-caliber 37mm's which were used for practice 
firing by Corregidor's big guns. These were dis- 
mounted from the gun tubes and turned over to 
the Marines. LtGen S. L. Howard interview by 
HistBr, G-3, HQMC, 260ct56, hereinafter cited 
as Howard Interview. 

" Maj T. E. Pulos Itr to CMC, 30Oct56. 



plane; the rest had to be left to accept 
their fate. 

On 11 March, the day before Mac Arthur 
and his party started the first leg of their 
journey to Australia, he created a new 
headquarters, Luzon Force, to control the 
operations on Bataan and appointed Gen- 
eral Wainwright to its command. On 20 
March, the War Department notified 
Wainwright of his promotion to lieutenant 
general and of the fact that he was to be 
commander of all forces remaining in the 
Philippines. To take the place of 
USAFFE, an area headquarters, United 
States Forces in the Philippines 
(USFIP), was created. 

To take his place on Bataan, Wainwright 
appointed the USAFFE Artillery Officer, 
Major General Edward P. King, Jr. King 
drew an unenviable task when he took over 
Luzon Force, for the volume of Japanese 
preparatory fire on Bataan and on the 
island forts indicated the start of a major 
effort. To meet this attack, King had 
troops who had already spent two weeks 
on a diet of % of a ration on top of two 
months of half rations; they were ready 
to fight but "with not enough food in their 
bellies to sustain a dog." ^» The USAFFE 
Surgeon General, on 18 February, had ac- 
counted Bataan's defenders as being only 
65% combat efficient as a result of "debil- 
ities due to malaria, dysentery, and general 
malnutrition." ^° These same men were 
now a month further along on the road to 
exhaustion and collapse and were destined 
to meet a fresh and vigorous enemy assault. 


The Fourteenth Arrn/y set 3 April as D- 
day for its reneAved offensive on Bataan, 
and General Homma foresaw "no reason 
why this attack should not succeed." ^^ He 
could well be confident since he had re- 
ceived the infantry replacements needed 
to rebuild the 16th Division and the 65th 
Brigade^ and he had been sent the J^th Di- 
vision from Shanghai. In addition, /m- 
perial Headquarters had allotted him a 
strongly reinforced infantry regiment 
from the 21st Division, originally slated 
for duty in Indo-China. His artillery 
strength had been more than doubled and 
now included far-ranging 240mm howitz- 
ers. Two heavy bomber regiments had 
been flown up from Malaya to increase ma- 
terially his mastery of the air. 

Once the enemy attack was launched, the 
pressure on Bataan's defenders was relent- 
less. In less than a week the issue had been 
decided. The physically weakened Ameri- 
cans and Filipinos tried desperately to 
stem the Japanese advance, but to no avail. 
By 7 April the last reserves had been com- 
mitted. A growing stream of dazed, dis- 
organized men, seeking to escape the in- 
cessant bombardment at the front and the 
onrushing enemy, crowded the roads and 
trails leading to Mariveles. Only isolated 
groups of soldiers still fought to hold the 
Japanese back from the tip of the penin- 
sula. Under these circumstances, General 
King decided to seek surrender terms. His 
aide recorded the situation in his diary : 

8th [April]. Wednesday. The army can not 
attack. It is impossible. Area is congested with 
stragglers . . . General King has ordered all 

'' Waimvrighfs Story, 76. The ration was cut 
to % on 2 March according to USAFFE-V8FIP 
Rept entry for that date. 

'" Quoted in Diary of Maj A. C. Tisdelle, Aide 
to MajGen King, entry of 18Feb42. 

-' Unless otherwise noted the material in this 
section is derived from U8AFFE-V8FIP Rept; 
nth Army Rept; Fall of the Philippines. 

=' IWi Army Rept, Appendix 4, 17. 



tanks thrown [blown?] and arms destroyed, and 
is going forward to contact the Japanese and try 
to avert a massacre. ^^ 

Near midnight on the 8th a severe earth- 
quake tremor was felt on Corregidor and 
Bataan, and soon thereafter the Mariveles 
harbor was shaking violently from man- 
made explosions, as King's orders to de- 
stroy all munitions dumps were carried 
out. To an observer on The Rock it seemed 
that : 

. . . the southern end of Bataan was a huge 
conflagration which resembled more than any- 
thing else a volcano in violent eruption . . . 
white hot pieces of metal from exploded shells 
and bombs shot skyward by the thousands in 
every conceivable direction. Various colored 
flares exploded in great numbers and charged off 
on crazy courses much the same as a sky rocket 
which has run wild on the ground." 

All night long the water between Bataan 
and Corregidor was lashed with falling 
debris and fragments from the explosions. 
Through this deadly shower a procession 

" Tisdelle, op. cit. 

" LCdr T. C. Parker, "The Epic of Corregidor- 
Bataan, December 24, 1941-May 4, 1942," U8NI 
Proceedings, January 1943, 18, hereinafter cited 
as Parker. 

of small craft dodged its way to the north 
dock of Bottomside. Everything that 
could float was pressed into use by frantic 
refugees. Some of the arrivals, however, 
such as the nurses from Bataan's hospitals, 
were under orders to report to Corregidor. 
Specific units that could strengthen The 
Rock's garrison, antiaircraft batteries and 
the 45th Philippine Scout Regiment, had 
also been called for. Only the AAA gun- 
ners from the Mariveles area, including 
the 4th Marines' Battery C, managed to 
escape. The Scout regiment was pre- 
vented from reaching the harbor in time 
by the jammed condition of the roads. 

By noon on 9 April, General King had 
found out that no terms would be given 
him; the Japanese demanded uncondi- 
tional surrender. With thousands of his 
men lying wounded and sick in open air 
general hospitals and all hope of success- 
ful resistance gone. King accepted the in- 
evitable and surrendered, asking only that 
his men be given fair treatment. The 
battle for Bataan had ended, and more 
than 75,000 gallant men began the first of 
more than a thousand days of brutal 


The Siege and Capture of Corregidor 


On 9 April the victorious Fourteenth 
Army paused on the shore of Bataan with 
its next target — Corregidor — dead center 
in its sights. Many enemy staff officers, 
both in Tokyo and on Luzon, wanted to 
launch an immediate amphibious attack, 
taking advantage of the army's success 
on Bataan. The dearth of landing craft 
in Manila Bay, however, effectively served 
to postpone the operation. Most of the 
Japanese landing barges and boats were 
located in Lingayen Gulf or Subic Bay 
and had to be moved past Corregidor's 
guns to the designated staging areas on 
the eastern coast of Bataan. (See Map 
8, Map Section) 

On the night of 14 April the first small 
group of boats slipped by The Rock, hug- 
ging Bataan's shore while the enemy 
shelled and bombed the island's north 
coast to prevent their discovery. ^ Because 
they were forced to follow this method of 
moving a few boats at a time and these 
only at night and behind a curtain of pro- 
tective fire, the Japanese took more than 
three weeks to assemble the necessary 
assault craft. 

The need for extreme caution in making 
the risky passage into Manila Bay was 
not the only factor which acted against 
rapid execution of the Japanese assault 

' Unless otherwise noted the material in this 
section is derived from l^th Army Reiit; Philip- 
pine AirOpsRec; Fall of the Philippines. 

' 4th MarJnl, 374. 

plan. In mid-April a severe outbreak of 
malaria in the ranks of the ^^A Division, 
Homma's chosen landing force, severely 
hampered attack preparations, but am- 
phibious training and rehearsals continued 
despite the temporary decrease in the 
division's effective strength. Emergency 
supplies of quinine tablets were flown to 
Luzon in time to check the spread of the 
disease and restore fighting trim. 

The Fourteenth Army was obsessed, with 
the need for deception and secrecy and 
stringent security measures were taken 
to conceal the preparations for the attack 
on Corregidor. A consistent effort was 
made to create the impression that Cavite 
Province was the Japanese amphibious 
"base and that Forts Frank and Drum were 
the targets. Landing craft maneuvered 
■off Cavite's shores while the army's air 
and artillery pounded the defenses of the 
southern islands. Two battalions of the 
16th Division feigned preparations for an 
attack on Frank and Drum, but there was 
little doubt at USFIP Headquarters that 
Corregidor was the primary Japanese 

Every day in April, starting with the 
day Bataan fell, an increasingly heavier 
concentration of enemy artillery pieces 
found firing positions in the peninsula's 
jungled hills. At least thirty-seven bat- 
teries, whose weapons ranged from 75mm 
mountain guns to 240mm howitzers, cov- 
ered Corregidor with a continuous pattern 
of fire that reached every position and 
knocked out the major portion of the 




island's defenses.^ Nine Japanese bomb- 
ing squadrons, capitalizing on the gradual 
weakening of antiaircraft fire, were over- 
head to add their bombardment to the at- 
tack preparation. 

The enemy J^th Division was reinforced 
for the assault with two independent en- 
gineer regiments to man the transport 
and support landing craft as well as 
a tank regiment and three mortar bat- 
talions to provide additional firepower. 
The actual landing operation was to be 
made in two stages with Colonel Gem- 
pachi Sato's 61st Infantry Regiment (two 
infantry battalions, a tank company, a 
mountain artillery battery, and mortar 
units) designated the initial assault force. 
Sato was to land his unit in successive 
waves, battalions abreast along the 
beaches between Infantry and Cavalry 
Points on the night of 5 May. After es- 
tablishing a beachhead, he was to send most 
of his men against Malinta Hill while the 
remainder of the regiment drove across the 
tail of the island to isolate and contain the 
defenders east of Infantry Point. The 
plan called for the 61st Regiment to be in 
possession of Malinta Hill by dawn, ready 
to support a second landing. 

Twenty-four hours after Sato's force 
landed, the division's main assault effort 
would strike beaches between Morrison 
and Battery Points, near James Ravine, 
and at the neck of the island. This second 
landing force, four heavily reinforced in- 
fantry battalions, would have the assist- 

' Many survivors and a number of accounts 
of this siege credit the Japanese with having 
as many as 400 artillery pieces firing on the 
fortified islands by 5 May. The figure of 37 
batteries (approximately 150 pieces) represents 
only the enemy artillery units listed in Hth Army 
Kept, 187 as part of the Corregidor attack 

ance of Sato's unit which was scheduled 
to make a concurrent attack against Ram- 
say Battery hill. Throughout the whole 
operation the artillery on Bataan, op- 
erating under army control, was to deliver 
preparatory and supporting fires, and in 
daylight hours the army's air squadrons 
were to fly close support missions. 

The ^th Division had three infantry 
battalions in reserve for its attack but did 
not expect that they would be needed. The 
Japanese were confident that their prepar- 
atory bombardment had knocked most of 
the fight out of Corregidor. Every ter- 
rain feature on the island was plotted and 
registered on artillery target maps and 
any signal for support from the assault 
forces would call down a smother of ac- 
curate fire on the defenders. The enemy 
felt certain that dusk of 7 May would see 
their assault troops in control of Cor- 


During the 27 days between the fall of 
Bataan and the assault on Corregidor, 
life on The Rock became a living hell. The 
men in the open gun pits and exposed 
beach defenses were subjected to an increas- 
ing rain of shells and bombs. It became 
virtually impossible to move about the 

* Unless otherwise noted the material in this 
section is derived from VSAFFE-V8FIP Rept; 
Moore Rept; 4th Mar Jnl; Hayes Rept; Capt C. 
B. Brook, USN, Personal Experiences 8Apr- 
6 May 56, n.d., hereinafter cited as Brook; Maj 
H. E. Dalness, USA, "The Operations of the 4th 
Battalion (Provisional) 4th Marine Regiment 
in the Final Counterattack in the Defense of 
Corregidor 5-6 May 1942," AdvInfOfE Course 
1949-50, The InfSch, Ft. Benning, Ga., herein- 
after cited as Dalness; Ferguson; Jenkins; 
IstLt O. E. Saalman, USA, Personal Experiences 
12Apr-6May42, n.d., hereinafter cited as 



AERIAL VIEW OF CORREGIDOR ISLAND showing in the foreground the area of the battle be- 
tween the Japanese landing force and the 4th Marines. ( SC 200883-S ) 

EFFECT OF JAPANESE BOMBARDMENT OF CORREGIDOR is shown in this photograph taken 
the day after the surrender near the main entrance to Malinta Hill. (SC 282343) 



island by daylight; enemy artillery spot- 
ters aloft in observation balloons on Ba- 
taan and in planes overhead had a clear 
view of their targets. The dense vegeta- 
tion which had once covered most of Cor- 
regidor was stripped away by blast and 
fragmentation to reveal the dispositions of 
Howard's command. The tunnels through 
Malinta Hill, their laterals crowded with 
headquarters installations and hospital 
beds, offered refuge for only a fraction of 
the 11,000-man garrison and the rest of 
the defenders had to stick it out with little 
hope of protection from the deadly down- 

Most of the escapees from Bataan were 
ordered to join the 4th Marines, thus add- 
ing 72 officers and 1,173 enlisted men to its 
strength between 9 and 12 April.'' The 
majority of the Army combat veterans, 
however, "were in such poor physical con- 
dition that they were incapable of even 
light work," * and had to be hospitalized. 
The mixed collection of infantry, artillery, 
aviation, and service personnel from both 
American and Philippine units assigned to 
the beach defense battalions was in little 
better shape than the men who had been 
committed to the hospital under Malinta 
Hill. The commander of 1/4's reserve, 
First Lieutenant Kobert F. Jenkins, Jr., 
who received a typical contingent of Ba- 
taan men to augment his small force com- 
mented that he : 

. . . had never seen men in such poor physical 
condition. Their clothing was ragged and 

' The former sergeant major of 2/4 believes 
that the regiment joined substantially more men 
than this figure which appears in the regimental 
journal. He recalls that the 2d Bn "picked up 
for rations, and on the crudest rolls, at least 600 
men" and believes that the other battalions did 
the same. Jackson. 

° Wainwriffht's Story, 87. 

stained from perspiration and dirt. Their gaunt, 
unshaven faces were strained and emaciated. 
Some of them were already suffering from beri- 
beri as a result of a starvation diet of rice for 
weeks. We did what we could for them and then 
put them to work on the beach defenses.' 

The sailors from Mariveles, mostly 
crewmen from the now-scuttled Canopus, 
were kept together and formed into a new 
276-man reserve battalion for the regi- 
ment, the 4th Battalion, 4th Marines.^ 
Not only was the designation of 4/4 un- 
usual, but so was its makeup and its per- 
sonnel. Only six Marines served in the 
battalion : its commander. Major Francis 
H. Williams, and five NCOs. The staff, 
company commanders, and platoon lead- 
ers were drawn from the nine Army and 18 
Navy officers assigned to assist Williams.* 
The four rifle companies were designated 
Q, R, S, and T, the highest lettered com- 
panies the men had ever heard of. An- 
other boast of the bluejackets turned Ma- 
rines was that they were "the highest paid 
battalion in the world, as most of the men 

' Jenkins, 13. 

' Most survivors of 4/4 refer to the battalion 
as having had approximately 500 men in its 
ranks. Strength breakdowns of the 4th Mar 
exist up through lMay42, however, and nowhere 
do they support the larger figure. The S-3 of 
4/4 is certain that the total strength of the bat- 
talion on 6 May was no more than 350 men. Maj 
O. E. Saalman Itr to CMC, 220ct56, hereinafter 
cited as Saalman 1956. 

° Survivors of 4/4 are unable to agree on the 
identity of the man who served as ExO of the 
battalion ; at least five Army or Navy ofiicers 
have been mentioned. In addition, the possibility 
that a Marine who was closely connected with 
4/4 was de facto ExO was brought out by one 
of the NCOs who recalls that "Major Williams 
always considered Gunner Joe Reardon [QMClk 
Joseph J. Reardon] as his Executive Officer and 
Adjutant." MSgt K. W. Mize Itr to CMC, lNov56. 



were petty officers of the upper pay 
grades." ^^ 

The new organization went into bivouac 
in Government Ravine as part of the regi- 
mental reserve. The reserve had here- 
tofore consisted of men from the 
Headquarters and Service Companies, re- 
inforced by Philippine Air cadets and 
Marines from Bataan. Major Max W. 
Schaeffer, who had replaced Major King 
as reserve commander, had organized this 
force of approximately 250 men into two 
tactical companies, O and P. Company 
O was commanded by Captain Robert 
Chambers, Jr. and Company P by Lieu- 
tenant Hogaboom; the platoons were led 
by Marine warrant officers and senior 

A good part of Schaeffer's men had pri- 
mary duties connected with regimental 
supply and administration, but each after- 
noon the companies assembled in the 
bivouac area where the troops were in- 
structed in basic infantry tactics and the 
employment of their weapons. Despite 
the constant interruptions of air raids and 
shellings, the Marines and Filipinos had 
a chance "to get acquainted with each 
other, familiarize themselves with each 
others' voices, and to learn [the] team- 
work" " so essential to effective combat 
operations. Frequently, Major Schaeffer 
conducted his company and platoon com- 
manders on reconnaissance of beach de- 
fenses so that the reserve leaders would be 
familiar with routes of approach and ter- 
rain in each sector in which they might 

While Schaeffer's unit had had some 
time to train before the Japanese stepped 
up their bombardment of the island in 

late March, Williams' battalion was organ- 
ized at the inception of the period of 
heaviest enemy fire and spent part of every 
day huddled in foxholes dug along the 
trail between Geary Point and Govern- 
ment Ravine. ^^ Any let-up in the bom- 
bardment would be the signal for small 
groups of men to gather around the Army 
officers and Marine NCOs for instruction 
in the use of their weapons and the tactics 
of small units. Rifles were zeroed in on 
floating debris in the bay and for most 
of the men this markmanship training was 
their first since Navy boot camp. When 
darkness limited Japanese shelling to 
harassment and interdiction fires, the 
sailors formed eager audiences for the 
Army Bataan veterans who gave them a 
resume of enemy battle tactics. Every 
man was dead serious, knowing that his 
chances for survival depended to a large 
extent upon how much he learned. "The 
chips were down; there was no horse- 
play." " 

To a very great extent the record of the 
4th Battalion in the fighting on Corregidor 
was a tribute to the inspirational leader- 
ship of its commander. During the try- 
ing period under enemy shellfire and 
bombing when the battalion's character 
was molded. Major Williams seemed to be 
omnipresent; wherever the bombardment 
was heaviest, he showed up to see how his 
men were weathering the storm. When 
on separate occasions Battery Crockett and 
then Battery Geary were hit and set afire, 
he led rescue parties from 4/4 into the 
resulting holocausts of flame, choking 
smoke, and exploding ammunition to 
rescue the wounded. He seemed to have 

" Brook, 1. 

" Ferguson, 10. 

" Dr. C. E. Chunn Itr to CMC, 12Nov56, herein- 
after cited as Chunn. 
" Dalness, 7. 



an utter disregard for his own safety in 
the face of any need for his presence. 
Survivors of his battalion agree with 
startling unanimity that he was a giant 
among men at a time when courage was 

Raw courage was a necessity on the forti- 
fied islands after Bataan's fall, since there 
was no defiladed position that could not 
be reached by Japanese 240mm howitzers 
firing from Cavite and Bataan. The 
bombers overhead, increasingly bold as 
gun after gun of the antiaircraft defenses 
was knocked out, came down lower to pin- 
point targets. Counterbattery and anti- 
aircraft fire silenced some enemy guns and 
accounted for a number of planes, but 
nothing seemed to halt the buildup of 
preparatory fires. 

On 28 April Howard issued a warning to 
his battalions that the next day would be 
a rough one. It was the Emperor's birth- 
day and the Japanese could be expected to 
"celebrate by unusual aerial and artillery 
bombardment." " The colonel's prophecy 
proved to be a true one, and on the 29th one 
observer noted that even "the kitchen sink 
came over." ^^ The birthday celebration 
marked the beginning of a period when the 
enemy bombarded the islands without let- 
up, day and night. The men manning the 
beach defenses of Corregidor's East Sector 
found it : 

. . . practically impossible to get any rest or 
to repair any damage to our positions and barbed 
wire. Our field telephone system was knocked 
out; our water supply was ruined (drinking 
water had to be hauled from the other end of the 
island in large powder cans) . . . Corregidor was 
enveloped in a cloud of smoke, dust, and the 
continuous roar of bursting shells and bombs. 

There were many more casualties than we had 
suffered in the previous Ave months." 

About three days prior to the Japanese 
landing, Lieutenant Colonel Beecher re- 
ported to Colonel Howard that defensive 
installations in the 1st Battalion's sector 

. . . practically destroyed. Very little defen- 
sive wire remained, tank traps constructed with 
great difficulty had been rendered useless, and all 
my weapons were in temporary emplacements as 
the original emplacements had been destroyed. 
I told Colonel Howard at this time that I was 
very dubious as to my ability to withstand a land- 
ing attack in force. Colonel Howard reported the 
facts to General Wainwright, who, according to 
Colonel Howard, said that he would never sur- 
render. I pointed out to Colonel Howard that I 
had said nothing about surrender but that I was 
merely reporting the facts as it was my duty 
to do." 

The increase in the fury of the Japanese 
bombardment with the coming of May, 
coupled with the frequent sightings of 
landing craft along the eastern shore of 
Bataan, clearly pointed to the imminence 
of an enemy landing attempt. The last 
successful effort to evacuate personnel 
from the island forts was made on the 
night of 3 May. The submarine Spearfish 
surfaced after dark outside the mine fields 
off Corregidor and took on a party of of- 
ficers and nurses who had been ordered 
out, as well as a load of important USFIP 
records and a roster of every person still 
alive on the islands.^* The 4th Marines 

" 4thMarJnl,S92. 
" Parker, 20. 

" Jenkins, 15-16. 

" BriGen C. T. Beecher Itr to Mr. G. J. Berry, 
17Mar50 (deposited by Capt G. J. Berry, USMCR, 
in the USMC Archives, 30Oct56). 

" Submarines were the beleaguered garrison's 
only contact with Allied bases outside the Phil- 
ippines during most of the siege. Although the 
subs brought in rations, antiaircraft ammunition, 
and medical supplies on scattered occasions, the 



sent out their regimental journal, its last 
entry, dated 2 May, the list of the five men 
who had been killed and the nine who had 
been wounded during the day's bombard- 

To one of the lucky few who got orders 
to leave on the Spearfish the receding is- 
land looked "beaten and burnt to a 
crisp." " In one day, 2 May, USFIP esti- 
mated that 12 240mm shells a minute had 
fallen on Corregidor during a five-hour 
period. On the same day the Japanese 
flew 55 sorties over the islands dropping 
12 1,000-pound, 45 500-pound, and 159 
200-pound bombs.^" The damage was ex- 
tensive. Battery Geary's eight 12-inch 
mortars were completely destroyed as was 
one of Battery Crockett's two 12-inch guns. 
The enemy fire also knocked out of action 
two more 12-inch mortars, a 3-inch gun, 
three searchlights, five 3-inch and three 
.50 caliber antiaircraft guns, and a height 
finder. Data transmission cables to the 
guns were cut in many places and all com- 
munication lines were damaged. The 
beach defenses lost four machine guns, a 
37mm, and a pillbox; barbed wire, mine 
fields, and antiboat obstacles were torn 

The logical landing points for an assault 
against Corregidor, the entire East Sector 
and the ravines that gave access to Topside 
and Middleside, received a special working 
over so steady and deadly that the effec- 
tiveness of the beach defenses was sharply 

amount that they could carry was only enough 
for stop-gap relief. For the interesting story of 
the diversified submarine actions in support of 
USAFFE-USFIP see, T. Roscoe, United States 
Submarine Operations in World War II (An- 
napolis: U. S. Naval Institute, 1949), 23-39. 

" Parker, 22. 

™ Philippine AirOpsRec, Plate 8. 

reduced. Casualties mounted as the men's 
foxholes, trenches, and shelters crumbled 
under the fire. Unit leaders checking the 
state of the defenses were especially vul- 
nerable to the fragments of steel which 
swept the ground bare. By the Japanese- 
appointed X-Day (5 May) the 1st Battal- 
ion had lost the commander of Company A, 
Major Harry C. Lang, and Captain Paul 
A. Brown, commanding Company B, had 
been hospitalized as a result of severe con- 
cussion suffered during an enemy bombing 
attack." Three Army officers attached to 
the Reserve Company, an officer of Com- 
pany B, and another of Company D had all 
been severely wounded. 

Despite the damage to defenses it had so 
laboriously constructed, the 4th Marines 
was ready, indeed almost eager, to meet a 
Japanese assault after days and weeks of 
absorbing punishment without a chance to 
strike back. On the eve of a battle which 
no one doubted was coming, the regiment 
was perhaps the most unusual Marine unit 
ever to take the field. From an under- 
strength two-battalion regiment of less 
than 800 Marine regulars it had grown 
until it mustered almost 4,000 officers and 
men drawn from all the services and 142 
different organizations.^ Its ranks con- 
tained 72 Marine officers and 1,368 enlisted 

" Beecher Itr to Berry, op. cit. 

" The last complete contemporary breakdown 
of strength of the 4th Mar by component units is 
contained in ith Mar J til, 390. It was corrected 
through 1 May. A slightly earlier list dated 
28Apr42, detached from the journal book but un- 
mistakably once part of it, has an interesting 
appendix which gives the units from which at- 
tached personnel originated. It shows that 26 
Navy, 104 American and Philippine Army, 9 
Philippine Scout, and 3 Philippine Constabulary 
organizations furnished men to the 4th Mar. 



Marines, 37 Navy officers ^^ and 848 blue- 
jackets, and 111 American and Philippine 
Air Corps, Army, Scout, and Constabulary 
officers with 1,455 of their men. 

The units that actually met the Japanese 
at the beaches — 1/4, 4/4, and the regimen- 
tal reserve — had such a varied makeup that 
it deserves to be recorded : ^* 

Service component 




4th Bn 









USMC & USMCR- ... 
















USN (MC & DC) 


USN . . - - - - 









Philippine Insular Navy 

Philippine Army Air Corps- 







Philippine Scouts .. _ .- 

Philippine Army. . . 

Philippine Constabulary. 


Totals . ... 










The area chosen by the Japanese for 
their initial assault, the 4th Marines' East 
Sector, was a shambles by nightfall on 5 

^ The five Marine officers, two Navy doctors, 
and 96 Marine enlisted men previously captured 
in China and on Bataan have been omitted from 
these figures. 

" The lMay42 listing of regimental strength 
does not indicate the tactical breakdown of Hq 
and SerCos into Cos O and P. The figures shown, 
therefore, include a number of regimental staff 
officers, probably two-thirds of the total, and a 
few enlisted men who did not serve in Maj 
Schaeffer's command. One officer and five en- 
listed men have been deducted from SerCo's 
USMC Strength and added to that of 4/4. 

"^ Unless otherwise noted the material in this 
section is derived from U8AFFE-U8FIP Rept; 
Moore Rept; 14th Army Rept; Howard Rept; 
MG H. M. Ferrell, Personal Experiences 5-6 
May42, n. d., hereinafter cited as Ferrell; 
Jenkins; H. W. Baldwin, "The Fourth Marines 
at Corregidor," MC Gazette, in 4 parts November 

May. Two days earlier the regimental in- 
telligence journal had noted that : 

There has been a distinct shifting of enemy 
artillery fire from inland targets to our beach 
defenses on the north side of Corregidor the 
past 24 hours." 

This concentration of fire continued and 
intensified, smashing the last vestiges of 
a coordinated and cohesive defensive zone 
and shaping 1/4's beach positions into an 
irregular series of strong points where a 
few machine guns and 37mm's were still 
in firing order. A pair of Philippine 
Scout-manned 75mm guns, located just 

1946-February 1947, hereinafter cited as Bald- 
win Narrative; Fall of the Philippines; K. Uno, 
Conegidor: Isle of Delusion (China: Press Bu- 
reau, Imperial Japanese Army Headquarters, 
September 1942) (located at OCMH), herein- 
after cited as Isle of Delusion. 

=" 4th Mar R-2 Jnl, 8Dec41-3May42, last entry. 



east of North Point, which had never re- 
vealed their position, also escaped the de- 
structive fires. Wire lines to command 
posts were ripped apart and could not be 
repaired; "command could be exercised 
and intelligence obtained only by use of 
foot messengers, which medium was un- 
certain under the heavy and continuous 
artillery and air bombardment." ^^ 

Along the northern side of the hogback 
ridge that traced its course from Malinta 
Hill to the bend in Corregidor's tail. Com- 
pany A and the reserves of 1/4 waited dog- 
gedly for the Japanese to come. There 
was no sharp division between unit defense 
sectors, and the men of the various units 
intermingled as the bombardment demol- 
ished prepared positions. Along the bat- 
tered base and sides of Malinta Hill, a 
special target for enemy fire, were the men 
of Lieutenant Jenkins' Reserve Company. 
Next to them, holding the shoreline up to 
Infantry Point, was a rifle platoon organ- 
ized from 1/4 's Headquarters Company; 
Captain Lewis H. Pickup, the company 
commander, held concurrent command of 
Company A, having taken over on the 
death of Major Lang. The 1st Platoon un- 
der First Lieutenant William F. Harris 
defended the beaches from Infantry to 
Cavalry Points, the landing site selected 
in Japanese pre-assault plans. Master 
Gunnery Sergeant John Mercurio's 2d Pla- 
toon's positions rimmed the gentle curve 
of land from Cavalry to North Point. Ex- 
tending from North Point to the tip of the 
island's tail were the foxholes and ma- 
chine-gun emplacements of First Sergeant 
Noble W. Well's 3d Platoon. 

Positions along the top of the steep 
southern face of the East Sector's domin- 
ant ridge were occupied by the platoons 

" VSAFFE-V8FIP Kept, 77. 

of Company B under First Lieutenant 
Alan S. Manning, who had taken over 
when Captain Brown was wounded.^* The 
machine guns and 37mm 's of Captain Noel 
O. Castle's Company D were emplaced in 
commanding positions along the beaches 
on both sides of the island ; the company's 
mortars were in firing positions near Ma- 
linta Hill. 

At about 2100 on 5 May, sensitive sound 
locators on Corregidor picked up the noise 
of many barges warming up their motors 
near Limay on Bataan's east coast. Warn- 
ing of an impending landing was flashed 
to responsible higher headquarters, but the 
lack of wire communication kept the word 
from reaching the men in the foxholes 
along the beaches of the East Sector. They 
did not need any additional advice of en- 
emy intentions anyway, since the whole 
regiment had been on an all-out alert every 
night for a month, momentarily expecting 
Japanese landing barges to loom out of 
the darkness. The men of 1/4 had with- 
stood some pretty stiff shellings, too, as 
the)' waited, but nothing to compare with 
the barrage that began falling on the 
beach defenses manned by Harris' 1st Pla- 
toon at about 2245. 

The Japanese had begun to deliver the 
short preparatory bombardment designed 
to cover the approach of Colonel Sato's 
assault waves which was called for in their 
operation plan. If Sato's boat groups 
adhered to their schedule they would ren- 
dezvous and head in for the beaches just as 
the artillery fire lifted and shifted to the 
west, walling off the landing area from 
American reinforcement efforts. The regi- 
ment would be ashore before the moon rose 
near midnight to give Corregidor's gun- 
ners a clear target. In two respects the 

"« LtCol R. F. Jenkins Itr to CMC, 30Oct56. 



plan miscarried, and for a while it was 
touch and go for the assault troops. 

The artillery shoot went off on schedule, 
but Sato's first waves, transporting most of 
his 1st Battalion^ were carried by an unex- 
pectedly strong incoming tide hundreds of 
yards to the east of the designated landing 
beaches. Guides in the oncoming craft 
were unable to recognize landmarks in the 
darkness, and from water level the tail of 
the island looked markedly uniform as 
smoke and dust raised by the shelling ob- 
scured the shoreline. The 61st RegimenVs 
2d Battalion, slated to follow close on the 
heels of the 1st, was delayed and disrupted 
by faulty boat handling and tide currents 
Until it came in well out of position and 
under the full light of the moon. 

When the Japanese preparatory fires 
lifted shortly after 2300, the troops along 
the East Sector beaches spotted the scat- 
tered landing craft of the 1st Battalion, 
61st heading in for the beaches at North 
Point. The few remaining searchlights 
illuminated the barges, and the island's 
tail erupted with fire. Enemy artillery 
knocked out the searchlights almost as soon 
as they showed themselves; but it made 
little difference, since streams of tracer 
bullets from beach defense machine guns 
furnished enough light for the Scout 75's 
near North Point and 1/4 's 37's to find tar- 
gets. A Japanese observer on Bataan de- 
scribed the resulting scene as "sheer massa- 
cre," ^' but the enemy 1st Battalion came in 
close enough behind its preparation to get 
a good portion of its men ashore. Although 
the Japanese infantrymen overwhelmed 
Mercurio's 2d Platoon, the fighting was 
fierce and the enemy casualties in the water 
and on the beach were heavy. Colonel 

' Quoted in Isle of Delusion, 17. 

Sato, who landed with the first waves, 
sorely needed his 2d Battalioii's strength. 

This straggling battalion which began 
heading shoreward about midnight suf- 
fered much more damage than the first 
waves. The remaining coast defense guns 
and mortars on Corregidor, backed up by 
the fire of Forts Hughes and Drum, 
churned the channel between Bataan and 
Corregidor into a surging froth, whipped 
by shell fragments and explosions. The 
moon's steady light revealed many direct 
hits on barges and showed heavily bur- 
dened enemy soldiers struggling in the 
water and sinking under the weight of 
their packs and equipment. Still, some 
men reached shore and Colonel Sato was 
able to organize a drive toward his objec- 
tive, Malinta Hill. 

Individual enemy soldiers and machine- 
gun crews infiltrated across Kindley Field 
and through the rubble of torn barbed 
wire, blasted trees, and crater-pocked 
ground to Denver Battery, a sandbagged 
antiaircraft gun position which stood on 
relatively high ground south of Cavalry 
Point. The American gunners, Avhose 
weapons were out of action as a result of 
the bombardment, were unable to beat back 
the encroaching Japanese who established 
themselves in a commanding position with 
fields of fire over the whole approach route 
to the landing beaches. Captain Pickup's 
first word that the Japanese had seized 
Denver Battery came when he sent one of 
Company D's weapons platoon leaders, 
Marine Gunner Harold M. Ferrell, to es- 
tablish contact with the battery's de- 
fenders. Ferrell and one of his men found 
the battery alive with enemy soldiers dig- 
ging in and setting up automatic weapons. 
Ferrell immediately went back to his de- 
fense area west of Infantry Point and 



brought up some men to establish a line 
"along the hogsback to prevent the enemy 
from coming down on the backs of the men 
on the beaches." ^° 

Pickup came up shortly after Gunner 
Ferrell got his men into position and con- 
sidered pulling Lieutenant Harris' platoon 
out of its beach defenses to launch an 
attack against the enemy. After a con- 
ference with Harris the company com- 
mander decided to leave the 1st Platoon 
in position. Japanese landing craft were 
still coming in, and the platoon's with- 
drawal would leave several hundred yards 
of beach open. The fact that enemy 
troops were ashore had been communi- 
cated to Lieutenant Colonel Beecher's CP 
just inside Malinta Tunnel's east entrance, 
and small groups of men, a squad or so at 
a time, were coming up to build on the 
line in front of Denver Battery. The 
enemy now fired his machine guns steadily, 
and intermittent but heavy shellfire struck 
all along the roads from Malinta to Den- 
ver. Casualties were severe throughout 
the area. 

By 0130 surviving elements of 1/4 on the 
eastern tip of the island were cut off com- 
pletely from the rest of the battalion. 
Beecher was forced to leave men in posi- 
tion on both shores west of Denver Battery 
to prevent the enemy landing behind his 
lines. All the men who could be spared 
from the beaches were being sent up to the 
defensive position astride the ridgeline 
just west of Denver, but the strength that 
could be assembled there amounted to little 
more than two platoons including a few 
Philippine Scouts from the silenced anti- 
boat guns in 1/4 's sector. No exact figures 
reveal how many Japanese were ashore at 

this time or how many casualties the 61st 
Infantry''s assault companies had suffered, 
but it was plain that the enemy at Denver 
Battery outnumbered the small force try- 
ing to contain them, and Japanese snipers 
and infiltrating groups soon began to crop 
up in the rear of Pickup's position. 

The situation clearly called for the com- 
mitment of additional men in the East Sec- 
tor. Colonel Howard had made provision 
for this soon after getting word of the 
landing attempt. He alerted Schaeffer's 
command of two companies first, but held 
off committing Williams' battalion until 
the situation clarified itself. There was 
no guarantee that the Japanese would ac- 
commodate the 4th Marines by landing all 
their troops in the East Sector; in fact, 
there was a general belief among the men 
manning the defenses which commanded 
the ravines leading to Topside that the 
East Sector landing was not the main 
effort and that the enemy would be coming 
in against West and Middle Sector 
beaches.^^ Complicating the entire prob- 
lem of command in the confused situation 
during the early morning hours of 6 May 
was the fact that only runners could get 
word of battle progress to Beecher's and 
Howard's CP. And any runner, or for 
that matter any man, who tried to make 
the 1,000-yard journey from the Denver 
line to the mouth of Malinta Tunnel stood 
a good chance of never completing his mis- 
sion. The area east of Malinta Hill was 
a killing ground as Schaeffer's men soon 
found out when they made their bid to 
reach Denver Battery. 

*> Ferrell, 1. 

" Hayes Rept, Statement of LCdr E. M. Wade, 
65. The general existence of this belief was 
questioned by one survivor. Jackson. 




In Government Ravine the 4th Marines' 
reserve companies saw and heard the ma- 
chine guns along the East Sector beaches 
hammering at the Japanese landing craft. 
Major Schaeffer's command was already 
standing by to move out, and near 2400 
Companies O and P filed down the trail 
and started for Malinta. There was little 
confusion, for the men had rehearsed their 
movements often. Crossing Bottomside 
by means of a tank trap which protected 
them from enemy shellfire, they moved into 
Malinta Tunnel where company and pla- 
toon commanders supervised the distribu- 
tion of machine-gun ammunition and gre- 
nades cached there for just such an emer- 
gency. Volunteers from the Navy and 
Marine headquarters installations joined 
the companies to serve as ammunitio;n 
carriers "although they were neither offi- 
cially or morally obligated to do so." ^ 

Major Schaeffer reported to Colonel 
Howard and received his instructions; he 
was to take his men out into the East Sec- 
tor and counterattack the Japanese posi- 
sion. At 0200 the companies began to 
move out of the oppressive heat and foul 
air of the crowded main tunnel onto the 
deeply cratered roads which led to Den- 
ver.^* Lieutenant Hogaboom's Company 

'■ Unless otherwise noted the material in this 
section is derived from USAFFE-U8FIP Kept; 
l^th Army Rept ; Howard Rept ; Ferguson; Fer- 
rell; Hogahoom; Jenkins; Baldwin Narrative ; 
Isle of Delusion; Fall of the Philippines. 

'^ Ferguson, 14. 

** By the time the Japanese landed, the only 
road into the East Sector was that which led 
through Malinta Tunnel. The road cut out of 
the side of the hill on the north had been com- 
pletely demolished, and Col Howard, looking for 
an alternate route of approach, had discovered 

P was in the lead, following the left fork 
of the road behind its guide, Captain 
Golland L. Clark, Jr., the 1st Battalion 
Adjutant. As the last platoon of the com- 
pany cleared the tunnel it was diverted to 
a vicious fire fight raging on the right of 
the Marine line by an officer who had come 
back seeking reinforcements. Several 
enemy machine guns had been set up near 
the base of a stone water tower forward 
of Denver Battery and to the right front 
of the Marine positions. The platoon, in 
common with most of the rest of the units 
that tried to reduce this strong point, was 
chopped to pieces by interlocking bands of 
machine-gun fire. 

On Clark's order, Hogaboom deployed 
his remaining two platoons in line of skir- 
mishers once they were well clear of the 
tunnel. The advancing line made contact 
with Lieutenant Harris and the remnants 
of Company A's 1st Platoon holding the 
left of the Denver defensive position and 
tied in with them. Hogaboom found that 
his right flank was open; Captain Cham- 
bers' Company O which was to have fol- 
lowed him out of the tunnel and come up 
on his right was not to be found. 

Chambers' men had left the tunnel all 
right, but almost immediately after the 
company column cleared the entrance 
bright flares were seen going up over the 
Japanese position. Chambers and his 1st 
Platoon leader. Quartermaster Clerk 
Frank W. Ferguson, concluding that the 
flares were a signal to the artillery on 
Bataan, passed the word along the line to 
look for the nearest shelter. The guess on 

shortly before the landing that enemy artillery 
had blown a deep, ravine-liUe depression in the 
southern circling road that rendered it impassable 
to organized troop movement. Howard Inter- 



the flares was right, and Ferguson's pla- 
toon was fortunate in taking its shelling in 
an area where the Japanese had provided 
deep bomb craters. The platoon came 
through with only eight casualties. As 
soon as the bombardment lifted, FergQson 
moved toward Denver until he was forced 
to deploy by heavy machine-gun and mor- 
tar fire. He looked for the 3d Platoon to 
come up on his right according to plan, but 
only its commander. Quartermaster Ser- 
geant John E. Haskin, and five men ap- 
peared, the rest had been lost in the shell- 
ing. Captain Chambers sent up the reserve 
platoon, which was in even worse shape, 
having been caught in the open near the 
tunnel entrance. Quartermaster Clerk 
Herman L. Snellings had only four sur- 
vivors alive and unwounded. 

Company O now contained but one pla- 
toon and had not yet made its attack. 

Major Schaeffer established control over 
the scattered groups of men from the 1st 
Battalion and the reserve and launched 
three separate counterattacks on the dug-in 
Japanese. Sometimes the men would get 
up the slopes leading to the battery gun 
pits, but they were always driven back, 
fewer in number each time. On the right 
flank. Sergeant Major John H. Sweeney 
and Sergeant Haskin took advantage of 
the water tower's battered elevation to hurl 
grenades down on the machine guns that 
were holding up the advance ; Haskin was 
killed trying to get more grenades up to 
Sweeney, and Sweeney was picked off 
after he had knocked out at least one of 
the guns. Ferguson, who knew and had 
served with both these long-time regulars, 
wrote their simple epitaph : 

They were very close friends in life and it was 
most fitting that they should go out together." 

Many close friends died that morning in 
the darkness and choking dust as the Jap- 
anese and the Americans and Filipinos 
faced each other from positions less than 
forty yards apart. Some men cut off be- 
hind the enemy lines still kept firing at 
occasional landing craft that were coming 
in to reinforce Sato. Hogaboom could see 
the tracers of a single .50 caliber and felt 
that "the bullets smacking into the armor 
of the barges sounded like rivet hammers 
rattling away." ^ Every movement of the 
Japanese boats which stood in number off- 
shore was counted as an attempt at land- 
ing, although many of them were impro- 
vised gunboats whose mission was protect- 
ing and supporting the landing craft. But 
detachments of Sato's force kept coming in 
all night, and one enemy lieutenant, prob- 
ably a member of one of the 61sfs support- 
ing units, gave a vivid description of the 
helpless feeling of the men in the barges 
as they were caught in Corregidor's fire: 

American high powered machine guns poured 
a stream of bullets on us from all directions. 
Rifle fire added to the hail of death. Our men 
who were huddled in the center of the boat were 
all either killed or wounded. Those who clung 
to the sides were hit by shells that pierced the 
steel plating. The boat had already sprung sev- 
eral leaks when we finally came within landing 
distance of Corregidor. Desperately I gave the 
signal and led the charge against the shore de- 
fenses. I don't remember how many men re- 
sponded. I know I heard only a small chorus. 
In that mad dash for shore many were drowned 
as they dropped into the water mortally 
wounded. Many were killed outright .... If 
it had not been for the fact that it was the dark 
hour before the dawn, pitch black, I doubt if any 
of us would be alive today to tell the story." 

However heavy the Japanese casualties 
were, they did not measurably weaken the 

' Ferguson, 18. 

'Hogaboom, 16. 

' Quoted in Isle of Delusion, 34. 



firepower of the Denver position. Each 
attack by Schaeffer's men thinned the Ma- 
rine line still more. Lost were officers and 
NCOs whose leadership was vital to the 
operations of mixed units such as those 
which held the Japanese at bay. Captain 
Castle of Company D was killed trying to 
silence a machine gun, and many small 
unit leaders who still held their place in 
line were badly wounded. The situation 
was so desperate that Colonel Howard 
could no longer hold his last reserves out 
of the action. He ordered the 4th Bat- 
talion to move into the East Sector and 
join the embattled defense line. 


Major Williams' 4th Battalion had been 
alerted early in the night's action, and he 
had ordered the issue of extra ammunition 
and grenades. At about 0100 he got the 
word to move the battalion into Malinta 
Tunnel and stand by. The sailors pro- 
ceeded cautiously down the south shore 
road, waited for an enemy barrage which 
was hitting in the dock area to lift, and 
then dashed across to the tunnel entrance. 
In the sweltering corridor the men pressed 
back against the walls as hundreds of 
casualties, walking wounded and litter 
cases, streamed in from the East Sector 
fighting. The hospital laterals were filled 
to overflowing, and the doctors, nurses, 
and corpsmen tended to the stricken men 
wherever they could find room to lay a 

^ Unless otherwise noted the material in this 
section Is derived from VSAFFE-VSFIP Kept; 
nth Army Rept; Howard Rept; Brook; Dalness; 
SSgt C. E. Downing, Personal Experiences 5-6- 
May42, n. d. ; Ferguson; Ferrell; Hogaboom; 
Jenkins; Baldxcin 'Narrative; Isle of Delusion; 
Fall of the Philippines. 

man down. At 0430, Colonel Howard 
ordered Williams to take his battalion out 
of the tunnel and attack the Japanese at 
Denver Battery. 

The companies moved out in column. 
About 500 yards out from Malinta they 
were caught in a heavy shelling that 
sharply reduced their strength and tempo- 
rarily scattered the men. The survivors 
reassembled and moved toward the fight- 
ing in line of skirmishers. Companies Q 
and R, commanded by two Army officers, 
Captains Paul C. Moore and Harold E. 
Dalness, respectively, moved in on the left 
to reinforce the scattered groups of rifle- 
men from Companies A and P who were 
trying to contain the Japanese in the 
broken ground north of Denver Battery. 
The battery position itself was assigned to 
Company T (Lieutenant Bethel B. Otter, 
USN) , and two platoons of Company S, *® 
originally designated the battalion reserve, 
were brought up on the extreme right 
where Lieutenant Edward N. Little, USN, 
was to try to silence the enemy machine 
guns near the water tower. The blue- 
jackets filled in the gaps along the line — 
wide gaps, for there was little that could 
be called a firm defensive line left — and 
joined the fire fight. 

The lack of adequate communications 
prevented Colonel Howard from exer- 
cising active tactical direction of the battle 
in the East Sector. The unit command- 
ers on the ground, first Captain Pickup, 
then Major Schaeffer, and finally Major 
Williams made the minute-to-minute de- 
cisions that close combat demanded. By 
the time Williams' battalion had reorgan- 
ized and moved up into the Marine for- 
ward positions, Schaeffer's command was 

' Saalman 1956. 



practically nonexistent. Williams, by 
mutual consent (Schaeffer was senior), 
took over command of the fighting since 
he was in a far better position to get the 
best effort out of his bluejackets when 
they attacked.*' 

At dawn Major Williams moved along 
the front, telling his officers to be ready 
to jump off at 0615. The company and 
platoon command posts were right up on 
the firing line and there were no reserves 
left; every officer and man still able to 
stand took part in the attack. On the left 
the Japanese were driven back 200-300 
yards before Williams sent a runner to 
check the advance of Moore and Dalness ; 
the right of the line had been unable to 
make more than a few yards before the 
withering fire of the Denver and water 
tower defenses drove the men to the deck. 
The left companies shifted toward Denver 
to close the gap that had opened while 
the men on the right tried to knock out 
the Japanese machine guns and mortars. 
Lieutenant Otter was killed while leading 
an attack, and his executive. Captain Cal- 
vin E. Chunn, took over; Chunn was 
wounded soon after as Company T charged 
a Japanese unit which was setting up a 
field piece near the water tower.^^ Lieu- 
tenant Little was hit in the chest and 
Williams sent a Philippine Scout officer. 
First Lieutenant Otis E. Saalman, to take 
over Company S. 

The Marine mortars of 1/4, 3-inch 
Stokes without sights, were not accurate 
enough to support Williams' attack. He 
had to order them to cease fire when stray 
rounds fell among his own men, who had 
closed to within grenade range of the Jap- 
anese. Robbed of the last supporting 

*" Mize Itr, op. cit. 
" Chunn. 

weapons that might have opened a breach 
in the Denver position, the attack stalled 
completely. Major Schaeffer sent War- 
rant Officer Ferguson, who had succeeded 
to command of Company O when Captain 
Chambers was wounded, to Colonel How- 
ard's CP to report the situation and re- 
quest reinforcements. Ferguson, like 
Schaeffer and many of the survivors of 1/4 
and the reserve, was a walking wounded 
case himself. By the time Ferguson got 
back through the enemy shelling to Ma- 
linta at 0900, Williams had received what 
few reinforcements Howard could muster. 
Captain Herman H. Hauck and 60 men of 
the 59th Coast Artillery, assigned by Gen- 
eral Moore to the 4th Marines, had come 
up and Williams sent them to the left 
flank to block Japanese snipers and ma- 
chine-gun crews infiltrating along the 
beaches into the rear areas. 

At about 0930 men on the north flank of 
the Marine line saw a couple of Japanese 
tanks coming off barges near Cavalry 
Point, a move that spelled the end on Cor- 
regidor. The tanks were in position to ad- 
vance within a half hour, and, just as the 
men in front of Denver Battery spotted 
them, enemy flares went up again and ar- 
tillery salvoes crashed down just forward 
of the Japanese position. Some men be- 
gan to fall back, and though Williams and 
the surviving leaders tried to halt the 
withdrawal, the shellfire prevented them 
from regaining control. At 1030 Williams 
sent a message to the units on the left flank 
to fall back to the ruins of a concrete trench 
which stood just forward of the entrance 
to Malinta Tunnel. The next thirty min- 
utes witnessed a scene of utter confusion 
as the Japanese opened up on the retreat- 
ing men with rifles, mortars, machine guns, 
and mountain howitzers. Flares signalled 



the artillery on Bataan to increase its fire, 
and a rolling barrage swung back and 
forth between Malinta and Denver, demol- 
ishing any semblance of order in the ranks 
of the men straining to reach the dubious 
shelter of the trench. "Dirt, rocks, trees, 
bodies, and debris literally filled the air," ^ 
and pitifully few men made it back to 

Williams, who was wounded, and 
roughly 150 officers and men, many of them 
also casualties, gathered in the trench ruins 
to make a stand. The Japanese were less 
than three hundred yards from their posi- 
tion and enemy tanks could be seen moving 
up to outflank their line on the right. The 
Marine major, who had been a tower of 
strength throughout the hopeless fight, 
went into the tunnel at 1130 to ask Howard 
for antitank guns and more men. But the 
battle was over : General Wainwright had 
made the decision to surrender. 


Colonel Howard had personally re- 
ported the landing of the Japanese tanks 
to General Wainwright at 1000, The 
USFIP commander, who had kept current 
on the situation in the East Sector 
throughout the night's fighting, made the 
fateful decision to surrender. He later 
related that "it was the terror that is vested 
in a tank that was the deciding factor," for 
he "thought of the havoc that even one of 
these could wreak if it nosed into the tun- 
nel, where lay our helpless wounded ..."** 
He did not believe, nor did any other oifi- 

" Dalness, 18. 

" Unless otherwise noted the material in this 
section is derived from VSAFFE-USFIP Kept; 
nth Army Rept; Howard Rept; Baldwin Narra- 
tive; Fall of the Philippines; Wainwright' s Story. 

** Wainwriffht'a Story, 119. 

cer he consulted, that the defenses outside 
Malinta could last more than the remain- 
ing hours of the day, and he set the hour of 
surrender for noon in order "to avoid the 
horrors which would have accrued had I 
let the fight go on until dark." ** 

The order to surrender was passed to the 
troops on Topside and Middleside along 
with instructions to destroy all weapons 
larger than .45 caliber. The sickened men 
of the 4th Marines' 2d and 3d Battalions, 
who had been forced to stand by helplessly 
as they heard and watched the battle to the 
east, carried the order even further, smash- 
ing their rifles against the rocks. Veterans 
of fighting in World War I and a dozen 
"banana wars" stood unashamedly crying 
as they were told they would have to sur- 
render. Inside Malinta, Colonel Howard 
ordered the regimental and national colors 
of the 4th Marines burned to prevent their 
falling into enemy hands. Two 1st Bat- 
talion officers. Captain Clark and Lieuten- 
ant Manning, a field music, and an inter- 
preter were selected to carry Wainwright's 
flag of truce to the Japanese. As the white 
flag was carried out of the tunnel, Major 
Williams ordered survivors of the East 
Sector fighting to move inside the hill and 
take shelter from the Japanese bombard- 
ment which still was falling. 

Captain Clark's party passed the last 
American outpost; the music sounded off 
and Manning waved a pole which bore a 
piece of sheeting. The enemy infantry- 
men, who had been given special instruc- 
tions regarding the reception of flags of 
truce, did not fire, and Clark was taken to 
the senior Japanese officer on the island 
who contacted Bataan and arranged for 
a parley on the peninsula with General 
Homma. When Wainwright, accompa- 

" Ibid., 186. 



denied by a few senior officers and aides, 
walked out of the tunnel and up the long 
slope toward Kindley Field, he saw dead 
and dying men on every hand, a grim 
record of the ferocity of the fighting in the 
past 12 hours. 

No complete figures exist for the casual- 
ties suffered by either side on 5-6 May; 
estimates of the Japanese losses range 
from 900 to 4,000.*« The strait between 
Bataan and Corregidor was heavily dotted 
with enemy bodies, and American pris- 
oners on Corregidor estimated that they 
helped collect and cremate the remains of 
hundreds of Japanese soldiers.*^ The de- 
tailed losses of the 4th Marines will prob- 
ably never be known because of the joint- 
service nature of the regiment at the time 
of battle and the scarcity of contemporary 
records. The casualties of Marines alone 
are known, however, and they may be con- 
sidered indicative of the fate of soldiers 
and sailors who served with them. In the 
whole Philippine campaign the regiment 

* Most American survivors of the battle men- 
tion that they heard from the Japanese later in 
prison camp that the enemy had suffered almost 
4,000 casualties in trying to take Corregidor. 
However, Japanese oflScers commenting on Dr. 
Morton's draft manuscript of The Fall of the 
Philippines wrote that the total casualties of the 
Japanese in the Corregidor operation between 
14Apr-7May42 were 903. MilHistSec, SS, GHQ, 
FEC, Comments of Former Japanese OflBcers Re- 
garding "The B"'all of the Philippines," 19Apr42 
(located at OCMH), Chap XXXI, 3. 

*' Saalman 1956. Saalman recalls having re- 
marked to MaJ Williams shortly after daybreak 
on 6 May, "I believe we could walk from Corregi- 
dor to Bataan over dead bodies." In light of the 
number of bodies that were collected and cre- 
mated, Saalman is convinced that the 903 figure 
supplied by the Japanese in commenting on the 
draft of The Fall of the Philippines reflects only 
enemy dead rather than total casualties. 
" See Appendix D for Marine casualties. 

had 315 officers and men killed, 15 missing 
in action presumed dead, and 357 men 
wounded;^* the great majority of these 
casualties occurred during the battle for 

The bloody battle for the island fortress 
did not end with Wainwright's decision to 
surrender. The Japanese went right ahead 
with their assault plan and preparatory 
bombardments, paying no heed to the 
white flags displayed on all the islands in 
the bay. Eighty-eight tons of bombs were 
dropped on 6 May, a good part of them 
after the surrender.** Wainwright, who 
had released his southern Philippine com- 
manders to MacArthur's control before he 
attempted to meet the enemy commander, 
tried to surrender only the fortified islands 
to the Japanese. He was rebuffed coldly 
by Homma's emissary and told that the 
Japanese knew that he was commander of 
all the forces in the Philippines and that 
they would not accept his surrender unless 
it meant the capitulation of every man in 
his command, everywhere in the islands. 
The American general, convinced that the 
Japanese would treat the men on the forti- 
fied islands as hostages, perhaps even mas- 
sacre them if the fighting continued in the 
south, finally acceded to the enemy demand 
and broadcast a surrender message at mid- 
night on 6 May to all his commanding 
officers. There was considerable dissension 
regarding this order, especially on islands 
where the Japanese had not made much 
effort to subdue the Philippine Army 
troops, but eventually most of the organ- 
ized units of USFIP came out of the hills 
to lay down their arms. Wainwright felt, 
as did most of his advisors at the time, that 
the Japanese were quite capable of slaugh- 
tering the men surrendered on the fortified 

' Philippine AirOpsRec, Plate 8. 



islands if he did not insure a complete 
surrender of all his forces. 

The struggle for control of Manila Bay 
finally ended on 7 May when the Japanese 
occupied the last of the island forts, but 
for most of the captured men "the fight 
for .life had just begun." ^^ Thousands 
succumbed in the next three years to brutal 
mistreatment, malnutrition, and disease in 
Japanese prison camps in the Philippines, 
in the enemy home islands, and in Man- 
churia. Two hundred and thirty-nine 
officers and men of the 4th Marine Regi- 
ment died in enemy hands. 


The battle for Corregidor was bitter 
and confused ; relatively few men survive 
who fought in the East Sector through the 
night and morning of 5-6 May 1942. 
Hundreds of well-trained infantrymen in 
positions within a mile or so of Malinta 
Hill were only spectators and auditors of 
the fighting. The poorest-trained ele- 
ments of the 4th Marines constituted the 
vital mobile reserve. On the surface and 
in hasty consideration it would seem that 
the tactics of the beach defense left much 
to be desired. 

Corregidor, however, was not a fortress 
with only one entrance. The beaches 
fronting the ravines defended by 2/4 and 
3/4 led directly to the island's major de- 
fensive installations. The threat of am- 
phibious assault existed all around the 
island's perimeter, but especially along the 
northern and western shores. The Japa- 
nese laid down preparatory fires all along 
the north side of the island, devoting as 
much attention to James Ravine and Bot- 
tomside west of Malinta as they did the 

eastern beaches. Until the night of 5 May 
there was no compelling reason to believe 
that the East Sector would draw the first 
assault. And even after the enemy landed 
at North Point the very present threat to 
western Corregidor existed and could not 
be ignored. To meet it, a number of Army 
units were alerted to back up the positions 
of 2/4 and 3/4.^1 

The problem which Colonel Howard 
faced of when, where, and in what strength 
to commit the reserves available to him 
was a classic one for commanders at all 
troop levels. If he committed all his re- 
serve at one time and in the area of 
greatest existing threat, he distinctly in- 
creased the vulnerability of other sectors 
to enemy attack. If he committed only 
part of his reserve and retained the capa- 
bility of reinforcing against further at- 
tacks, he stood the chance of not using 
enough men to have a decisive effect in 
any sector. The decision to commit the 
reserve piecemeal reflected the regiment's 
estimate of the enemy's capabilities and 
intentions in light of their actions.^^ The 
Japanese, although opposed by a rela- 
tively small force, did not or could not 
vigorously pursue their advance after 
reaching the Denver position. The con- 
tinued presence of numerous small craft 
off Corregidor's north shore indicated a 
possible, even probable, early attempt at 
a second landing. Under these circum- 
stances the East Sector assault might well 
be a secondary effort which had stalled, 
with the enemy's main attack still to come. 
Actually this was the Japanese plan, with 
the difference that the second landing was 
to follow the first after a day's interval 

™ Jenkins, 19. 

" Col F. P. Pyzick Itr to CMC, 30Oct56. 
^^ Howard Interview. 



rather than as soon as the Marines 

In large part the 4th Marines' reserve 
strength was already committed on 5 May. 
The Japanese preparatory fires, especially 
those which were laid on areas in plain 
sight of Bataan, made movement by any 
body of troops extremely difficult — witness 
the fate of Company O. The bombard- 
ment had the effect of tying the regiment 
to its defenses. The trained infantrymen 
in its ranks were kept where they could do 
the most to bolster the crucial beach posi- 
tions.^^ If any sizable number of these 
men had been withdrawn from the beaches 
to form a reserve, it is questionable whether 
the remaining' men could have withstood 
any enemy assault. Once the Japanese be- 
gan to bombard Corregidor in earnest 
there was no such thing as a strong beach 
defense position ; the very fury of the bom- 
bardment, destroying as it did most of the 
prepared defenses and demolishing the 
major supporting weapons, placed a high 


premium on having the best infantrymen 
at the point where their value would be 
greatest — the beaches. 

The fall of Corregidor was inevitable; 
the garrison simply did not have enough 
food to hold out until relief could arrive. 
Although the enemy, primarily for pres- 
tige and propaganda reasons, choose to as- 
sault the island, they could easily have 
starved its defenders into submission. 
When the Japanese did make their attack 
they paid a high price for their haste, 
but extracted as great a one from the de- 
fenders. In the immediate tactical sense, 
however, the enemy artillery was the vic- 
tor in the siege and fall of Corregidor ; no 
defending force could have withstood its 
devastatingly accurate bombardment. 

Although it was a defeat, the battle of 
Corregidor is marked down in the annals 
of the 4th Marines as a fight to be proud 
of. Those who fought and died in its 
ranks, whatever their service of origin, 
were, if only for a brief moment, Cor- 
regidor Marines. 


Decision at Midway 


Setting the Stage: 
Early Naval Operations^ 

Early in January 1942 the U. S. Pacific 
Fleet was looking for a way to strike back 
at the Japanese ; and advocates of the fast 
carrier force, believing their case ironically 
had been proven by the Japanese at Pearl 
Harbor, were ready to test their theories.^ 
But first there were some fences to mend. 
The all-important communications chain 
to Australia and New Zealand was rather 
tenuous, and the bulk of the Navy had to 
be used for escort duty until reinforce- 
ments could be put ashore to bolster some, 
of the stations along the route. 

One of the most important links in this 
communication chain was Samoa. The 
worst was feared for this area when the 

' Unless otherwise noted the material in Part 
V is derived from Rising Sun in the Pacific; S. E. 
Morison, Coral Sea, Midway and Submarine Ac- 
tions, May 19^2-August 1942: History of United 
States Naval Operations in World War II (Bos- 
ton : Little, Brown and Company, 1949), here- 
inafter cited as Coral Sea and Midway; U. S. d 
Sea Power; M. Fuchida and M. Okumiya, Mid- 
way: The Battle That Doomed Japan (Annapo- 
lis : U. S. Naval Institute, 1955), hereinafter cited 
as Battle That Doomed Japan; Marines at Mid- 
nmy; "The Japanese Story of the Battle of Mid- 
way," ONI Review, May 1947, hereinafter cited 
as ONI Review. 

■ 2 "There was still much difference of opinion 
[about the effectiveness of the carrier striking 
force] until 7 Dec 1941 when the Japanese attack 
took the controversy out of the laboratory class 
. . . Japan knifed us with our own invention." 
Capt Miles R. Browning. USN, "The Fast Carrier 
Force," MC Gazette, June 1946, 19. 

Pago Pago naval station was shelled by 
a Japanese submarine on 11 January while 
the 2d Marine Brigade (composed for the 
most part of the 8th Marines and the 2d 
Defense Battalion) still was en route to the 
islands from San Diego. But on 23 Jan- 
uary after the Marines' four transports 
and one fleet cargo vessel were delivered 
safely to Samoa, Vice Admiral William 
F. Halsey's Enterprise force, together 
with a new fast carrier force commanded 
by Rear Admiral Frank Jack Fletcher 
and formed around the Yorktown^ were re- 
leased for the raiding actions that the fleet 
was anxious to launch. 

While the 2d Marine Brigade unloaded 
at Samoa on 23 January, the Japanese 
landed far to the west at Rabaul where 
the small Australian garrison was quickly 
overrun. Although the importance of 
Rabaul to the Japanese was not realized 
at once, it was soon clear that from the 
Bismarcks the enemy could launch an at- 
tack through the Coral Sea toward Aus- 
tralia and New Zealand. This threat 
tended to increase rather than diminish 
the danger to Samoa. It was reasoned 
that a Japanese attack there would precede 
a strike at Australia or New Zealand to 
block U. S. assistance to the Anzac areas. 
Japanese occupation of Makin Island in 
the Gilberts seemed to point toward 
Samoa, and naval commanders held that 
the best insurance against subsequent 






moves would be a raid against the Mar- 
shalls, from which much of this Japanese 
action was mounted. Halsey's Enterprise 
force therefore set out to strike Wotje 
and Maleolap, seaplane bases in the Mar- 
shalls, while Fletcher prepared to attack 
Mili and Jaluit (also in the Marshalls) 
plus Makin with his Yorktown group. 
(See Map 9) 

A submarine reconnaissance found the 
Marshalls lightly defended and spot- 
ted the largest concentration of Japanese 
planes and ships at Kwajalein Atoll in 
the center of the island group. Halsey 
decided to add this choice target to his 
list, and for the missions he divided Task 
Force 8 into three groups. The Enterprise 
with three destroyers would strike Wotje, 
Maloelap, and Kwajalein; Rear Admiral 
Raymond A. Spruance with the cruisers 
N orthanvpton and Salt Lake City plus one 
destroyer would bombard Wotje; and 
Captain Thomas M. Shock in USS 
Chester^ and with two destroyers, would 
shell Maloelap. The three southern atolls 
in the Marshalls group and Makin in the 
northern Gilberts would be attended to by 
Fletcher in the Yorktown with his inde- 
pendent command (TF 17) made up of the 
cruisers Louisville and St. Louis and four 

The twin attacks struck on 1 February. 
Halsey began launching at 0443 under a 
full moon when his carrier was just 36 
miles from Wotje. Kwajalein, the main 
objective was 155 miles away. Nine tor- 
pedo bombers and 37 dive bombers led off 
the attack, the SBD's striking Roi air base 
on the northern end of the atoll and the tor- 
pedo bombers hitting Kwajalein Island 
across the lagoon. 

At Kwajalein the hunting was better; 
but in spite of the fact that there was no 

fighter opposition, and that the reports 
brought back by pilots were enthusiastic, 
damage to the Japanese installations and 
shipping was slight. Five Wildcats shot 
down two Japanese planes over Maloelap, 
and nine SBD's that returned from Roi 
later sortied again and damaged some air- 
field installations. The surface bombard- 
ment, too, was disappointing, and the 
bombardment flagship, Chester., took a 
light bomb through her main deck and lost 
eight men killed and eleven wounded. 

To the south, Fletcher had bad luck over 
Jaluit when his fliers found their targets 
concealed by thunder showers. Two Jap- 
anese ships off Jabor Town were hit, but 
not sunk, and the damage ashore was 
slight. A mine layer was hit at Makin, 
and damage at Mili was also slight. 

Similar actions were continued in other 
areas of the Pacific to harass the Japanese 
and to provide at least one outlet for efforts 
to fight back at the enemy when the news 
from all other fronts was gloomy. Most 
notable were strikes in early March against 
Wake and Marcus Island, and the daring 
raid by planes of Admiral Wilson Brown's 
task force over New Guinea's 15,000-foot 
Owen Stanley Mountains to hit the Jap- 
anese newly moved into Lae and Salamaua. 
But in all cases actual damage to the 
enemy still failed to measure up to ex- 
pectations, much less to the reports turned 
in by overenthusiastic aviators. 

Most audacious and unorthodox of the 
attacks, of course, was that which launched 
Lieutenant Colonel James H. Doolittle and 
his Army raiders from the Hornefs deck 
to the 18 April Tokyo raid. Planned as 
"something really spectacular" — a proper 
retaliation for Pearl Harbor — the raid was 
designed more for its dramatic impact 
upon morale than for any other purpose. 
In that it was highly successful. 



AN ARMY B-25, one of Doolittle's Raiders, takes off from the deck of the carrier Hornet to participate in 
the first U. S. air raid on Japan. ( USN 41197) 

JAPANESE CARRIER SHOHO, dead in the water and smoking from repeated bomb and torpedo hits, 
was sunk by carrier planes in the Coral Sea Battle. (USN 17026) 



After security-shrouded planning and 
training, Doolittle's 16 B-25's left San 
Francisco on 2 April 1942 on board the 
Hornet which was escorted by cruisers 
Yvrwennes and Nashville, four destroyers, 
and an oiler. After a 13 April rendezvous 
with the Enterprise of Halsey's TF 16, the 
raiding party continued along the north- 
ern route toward the Japanese home 
islands.^ Enemy picket boats sighted the 
convoy when it was more than 100 miles 
short of the intended launching range, 
and, with Doolittle's concurrence, Halsey 
launched the fliers at 0725 on 18 April 
while the Hornet bucked in a heavy sea 
668 miles from the Imperial Palace in 
central Tokyo.* 

Much of the raid's anticipated shock 
effect on Tokyo was lost by the coincidence 
of Doolittle's arrival over the city at about 
noon just as a Japanese air raid drill was 
completed. The Japanese, confused by 
the attack which followed their own 
maneuvers, offered only light opposition 
to the B-25's which skimmed the city at 
treetop level to drop their bombs on mili- 
tary targets. One plane which struck 
Kobe received no opposition, although two 
others over Nagoya and Osaka drew heavy 
fire from antiaircraft batteries; but none 
was lost over Japan. 

Halsey managed to retire from the 
launching area with little difficulty, and 

' Plans called for the bombers to land on 
friendly Chinese fields some 1,093 miles from 
Tokyo, and completion of this trip for the planes 
loaded initially with four 500-pound bombs and 
1,140 gallons of gasoline required that they be 
launched within 500 miles of Tokyo. 

* Although the picket boats were prompt with 
a warning, Japanese interception attempts were 
tardy. It was assumed that Navy planes, with 
a shorter range than B-25's, were on the carrier, 
and that the force could not strike Japan until 
the ships steamed for another day. 

both carriers returned to Pearl Harbor on 
25 April. Although the raid did more to 
boost American morale than it did to 
damage Japanese military installations, 
more practical results came later. It al- 
lowed a Japanese military group which 
favored a further expansion of their terri- 
torial gains to begin execution of these 
ambitious plans, and this expansion effort 
led directly to the Battle of Midway 
which ". . . alone was well worth the ef- 
fort put into this operation by . . . [those] 
. . . who had volunteered to help even the 
score." ^ 

The first of Japan's planned expansion 
moves in the spring of 1942 aimed for 
control of the Coral Sea through seizure 
of the Southern Solomon Islands and 
Port Moresby on New Guinea as bases 
from which to knock out growing Allied 
air power in northeastern Australia. 
Seizure of New Caledonia, planned as part 
of the third step in the second major series 
of offensives, would complete encircle- 
ment of the Coral Sea. This would leave 
the U. S. communications route to the 
Anzac area dangling useless at the Samoan 
Islands; and later Japanese advances 
would push the U. S. Pacific Fleet back to 
Pearl Harbor and perhaps even to the 
west coast. 

Characteristically, the Japanese plan 
called for an almost impossible degree of 
timing and coordination. It depended for 
success largely on surprise and on the 
U. S. forces behaving just as the Japanese 
hoped that they would. But this second 
element was largely corollary to the first, 
and, when surprise failed, the Japanese 
were shocked to discover that the U. S. 
fleet did not follow the script. 

' Rising Sun in the Pacific, 398. 



The Japanese anticipated resistance 
from a U. S. carrier task force known to be 
lurking somewhere to the south or south- 
east, but they expected to corner this force 
in the eastern Coral Sea with a pincers 
movement of carrier task forces of their 
own. Vice Admiral Takeo Takagi would 
skirt to the east of the Solomons with his 
Striking Force of heavy carriers Shokaku 
and ZmkaJcu and move in on the U. S. ships 
from that direction, while Rear Admiral 
Aritomo Goto's Covering Force built 
around the light carrier Shoho would close 
in from the northwest. Destruction of the 
northeast Australian airfields would fol- 
low this fatal pinch of the U. S. fleet, and 
then the Port Moresby Invasion Group 
could ply the southeastern coast of New 
Guinea with impunity. 

But Japanese overconfidence enabled 
U. S. intelligence to diagnose this opera- 
tion in advance, and Fletcher's Task Force 
17 had steamed into the Coral Sea where he 
all but completed refueling before the first 
Japanese elements sortied from Rabaul. 
On 4 May Fletcher's Lexington, Yorktown, 
screening ships, and support vessels were 
joined by the combined Australian- Ameri- 
can surface force under Rear Admiral J. C. 
Grace of the Royal Navy. 

On the previous day the Japanese had 
started their operation (which they called 
"Mo") like any other routine land grab. 
A suitable invasion force, adequately sup- 
ported, moved into the Southern Solomons, 
seized Tulagi without opposition® and 
promptly began setting up a seaplane base. 
There Fletcher's planes startled them next 

day with several powerful air strikes on 
the new garrison and on the Japanese ships 
still in the area. The U. S. carrier planes 
struck virtually unopposed,^ but they 
caused little damage in proportion to the 
energy and ammunition they expended. 
This startling deviation from the "Mo" 
script caused the Japanese to initiate the 
remaining steps of the operation without 
further delay. 

The Battle of the Coral Sea proved the 
first major naval engagement in history 
where opposing surface forces neither saw 
nor fired at each other.* Although both 
were eager to join battle, combat intelli- 
gence was so poor and aerial reconnais- 
sance so hampered by shifting weather 
fronts that three days passed before the 
main forces found each other. But other 
things they did find led to a series of events 
on the 7th that might be described as a 
comedy of errors, although there was noth- 
ing particularly comical about them to 
those involved. 

Early that morning an over-enthusiastic 
Japanese search pilot brought Takagi's en- 
tire striking air power down on the U. S. 
fleet tanker Neosho and her lone convoying 
destroyer, the USS Sims, by reporting 
them as a carrier and heavy cruiser re- 
spectively. This overwhelming attack 
sank the Sims and so damaged the Neosho 

"Tulagi, with which U. S. forces were to get 
better acquainted, was the capital of the British 
Solomon Islands Protectorate. Officials and 
such garrison as existed had been amply fore- 
warned and evacuated several days earlier. See 
Part VI of this volume. 

' Experience to date had indicated to the Jap- 
anese that one of their landings constituted a 
fait accompli which no enemy would dare dis- 
pute, and the naval force supporting the landing 
had departed in order to get on with the war. 
Takagi's powerful Carrier Striking Force at this 
time lay north of Bougainville to keep beyond 
the range of Allied air search. 

" "So many mistakes were made on both sides 
in this new mode of fighting that it might be 
called the Battle of Naval Errors." Coral Sea 
and Midtcay, 63. 



that she had to be destroyed four days 

Not to be outdone, the Americans reacted 
similarly a short time later to a scout 
plane's report of two Japanese carriers and 
four cruisers north of the Louisiades. Ac- 
tually these craft were a subordinate enemy 
task group consisting of two old light 
cruisers and three converted gunboats. 
But more by good luck than good manage- 
ment, the attacking planes investigating 
the report sighted the Japanese Covering 
Force, then protecting the left flank of the 
Port Moresby Invasion Group, and con- 
centrated on the Shoho to the virtual ex- 
clusion of her consorts. Against 93 air- 
craft of all types, the lone light carrier had 
no more chance than Task Force Neosho- 
Sinis had against the Japanese, and her 
demise prompted the morale-boosting 
phrase, "Scratch one flattop!"^ 

As a result of these alarms and excur- 
sions, both commanding admirals had 
missed each other once again. By mid- 
afternoon, however, Takagi had a pretty 
good idea of the U. S. carriers' location 
and, shortly before nightfall, dispatched a 
bomber-torpedo strike against Fletcher. 
Thanks to a heavy weather front, these 
planes failed to find their target, and 
American combat air patrol intercepted 
them on their attempted return. In the 
confusion of dogfights, several Japanese 
pilots lost direction in the gathering dark- 
ness and made the error of attempting to 
land on the Yorktown}^ 

° "Scratch one flattop ! Dixon to Carrier, 
Scratch one flattop!" Voice radio report LCdr 
R. E. Dixon to USS Lexington, quoted in Coral 
Sea and Midway, 42. Action against the Shoho 
was U. S. Navy Air's first attack on an enemy 

" Six planes in two groups of three each. Al- 
though they were recognized and fired on, all but 

Early the following morning, U. S. 
search planes finally located the Japanese 
carriers at about the time the Japanese re- 
discovered the U. S. flattops. At last the 
stage was set for the big show. 

Loss of the Shoho had cut the Japanese 
down to size. The opponents who slugged 
it out on 8 May 1942 were evenly matched, 
physically and morally, to a degree rarely 
found in warfare, afloat or ashore." How- 
ever, at the time the battle developed, the 
Japanese enjoyed the great tactical advan- 
tage of having their position shrouded by 
the same heavy weather front that had 
covered the U. S. carriers the previous 
afternoon, while Fletcher's force lay in 
clear tropical sunlight where it could be 
seen for many miles from aloft. 

The attacking aircraft of both parties 
struck their enemy at nearly the same time 
(approximately 1100), passing each other 
en route.^^ The two Japanese carriers and 
their respective escorts lay about ten miles 
apart. As the Yorktown's planes orbited 
over the target preparatory to the attack, 
the Zuikaku and her screening force dis- 
appeared into a rain squall and were seen 

one escaped. In this action the Japanese lost 9 
planes in combat and 11 attempting night land- 
ings without benefit of homing devices, against 
U. S. loss of 2 fighters. 

" A lucid summary of the several factors in- 
volved occurs in Coral Sea and Midway, 48. 
Fletcher's potential marked advantage in surface 
screening strength had been dissipated when, 
early on 7 May, he had dispatched Grace's force of 
cruisers and destroyers on a dash westward to 
intercept the enemy transport convoy expected to 
round the southeastern tip of New Guinea the 
next morning en route to attack Port Moresby. 
For analysis of this perhaps ill-judged move and 
its results, see IMd., 37-39. 

" "The story current shortly after the battle, 
that the Japanese and American planes sighted 
but paid no attention to each other when passing 
on opposite courses, is not true." IMd., 52n. 



no more during the brief action that fol- 
lowed, thereby escaping' damage. So all 
U. S. planes that reached the scene concen- 
trated on the Shokaku, but with disap- 
pointing results. 

The Yorktown's torpedo bombers went 
in first, low and covered by fighters. But 
faulty technique and the wretched quality 
of U. S. torpedoes at this stage of the war 
combined to make this attack wholly in- 
effective: hits (if any) proved to be duds, 
the pilots launched at excessive ranges, 
and the torpedoes traveled so slowly that 
vessels unable to dodge had only to out- 
run them. The dive bombers, following 
closely, scored only two direct hits. But 
one of these so damaged the Shakaku's 
flight deck that she could no longer 
launch planes, although she still was 
capable of recovering them. Many of the 
Lexington planes, taking off ten minutes 
after those from Yorktown, got lost in the 
overcast and never found their targets. 
Those that did attack made the same mis- 
takes the Yorktown fliers committed. The 
torpedoes proved wholly ineffective, and 
the damaging bomb hit on the Shokaku 
was something less than lethal despite the 
pilot's enthusiastic report that she was 
"settling fast." " 

The Japanese did considerably better, 
thanks to vastly superior torpedoes and 
launching techniques. Two of these 
powerful "fish" ripped great holes in the 
Lexington's port side, and she sustained 
two direct bomb hits plus numerous near 
misses that sprang plates. The more 
maneuverable Yorktown dodged all of the 
torpedoes aimed at her and escaped all 
but one of the bombs. But this was an 
800-pounder, and it exploded with such a 
spectacular display of flame and smoke 

that the Japanese pilots may be excused 
their claim that they had sunk her. 

These events made up the Battle of the 
Coral Sea. It was all over by 1140. 

Preoccupation of both forces with the 
flattops left opposing escort vessels un- 
scathed, although the Japanese claimed to 
have left burning "one battleship or 
cruiser." " The Americans had sustained 
far the heavier damage and casualties, but 
had inflicted the greater tactical blow in 
knocking the Shokaku out of further of- 
fensive action while both U. S. carriers 
still were operational. Even the crippled 
Lexington had put out fires, shored up 
torpedo damage, and was capable of sus- 
taining 26-knot speed and conducting 
nearly normal flight operations an hour 
after the battle ended. 

The Japanese had lost the greater num- 
ber of planes : 43 from all causes against 
33 for the Americans. Their command, 
accepting at face value the ecstatic reports 
of their pilots that they had sunk both U. S. 
carriers, started the beat-up Shokaku for 
home, and in the afternoon commenced 
withdrawal from the area on orders from 
Rabaul. Admiral Takagi concurred with 
higher authority that it would be unwise 
to risk the vulnerable transport convoy in 
the narrowing waters of the western Coral 
Sea in face of the Allies' Australian air- 
fields under cover of the whittled-down 
air complement of the single operational 
carrier. So the Port Moresby Invasion 
Group was ordered back to Rabaul. 

But the final, tragic act of the drama re- 
mained. The gallant old Lexington^ her 
wounds patched up and apparently fit to 
return to Pearl Harbor for permanent re- 
pairs, was suddenly racked by a terrible 

" lUd., 51. 

' Ihid., 56. 



explosion. This resulted only indirectly 
from enemy action : released gasoline 
fumes were ignited by sparks from a gen- 
erator someone had carelessly left running. 
This set off what amounted to a chain I'e- 
action. The best efforts of her crew availed 
nothing, and at 1707 her skipper gave the 
order to abandon ship. This movement 
was carried out in the best order, without 
the loss of a man. At about 2000, nearly 
nine hours after the Japanese had with- 
drawn from the battle, torpedoes of her 

own escort put her under the waves 

Loss of the Lexington gave tactical vic- 
tory to the Japanese. But by thwarting 
the invasion of Port Moresby, principal 
objective of the entire operation, the 
United States won strategic victory. At 
the time the enemy regarded this merely 
a postponement of their invasion plans; 
but events would prove that no Japanese 
seaborne invasion ever would near Port 
Moresby again. 


Japanese Plans: Toward Midway 
and the North Pacific 

Apparently ignoring this setback in the 
Coral Sea, Japan next turned toward the 
Central and North Pacific to launch the 
second complicated operation on her sched- 
ule. Admiral Yamamoto's two-pronged 
thrust at Midway and the Aleutians would 
automatically wipe clean the Coral Sea 
reverses and extend the outer perimeter 
of defense a safer distance from the home 
islands. And in the bargain, Yamamoto 
hoped, these attacks would lure forth the 
remainder of the U. S. fleet so that he 
could finish off the job he started on 7 

The admiral accepted his aviators' re- 
ports that they had destroyed both U. S. 
carriers in the Coral Sea, and he there- 
fore reasoned that the U. S. could bring 
no more than two flattops against him any- 
where in the Pacific. Actually the Pearl 
Harbor yard had put the Yorktown back 
into operation in less than 48 hours so that 
the U. S. had three carriers, including the 
Enterprise and the Hornet. But against 
these Yamamoto had seven, and four sea- 
plane carriers as well. His force also con- 
tained 11 battleships, including three of 
the latest type.^ The U. S. had no battle- 

' Battle That Doomed Japan, Chaps 4 and 5, 
passim. The Aleutians phase was intended only 
as a diversion and to protect the northern flank 
of the Midway thrust, the plan being to with- 
draw the landing troops in September. Ihid., 79. 

2 Adm Yamamoto flew bis flag in the Yamato, 
the largest, fastest, and most heavily armed (18" 
guns) ship in the world. 

ships — or at least none in position for this 
upcoming battle.^ And Yamamoto also 
had a substantial edge over the U. S. in 
cruisers, destroyers, and submarines. But 
the Japanese admiral squandered this lop- 
sided advantage by dispersing his armada 
in widely scattered groups and opened 
himself for defeat in detail by the inferior 
U. S. Pacific Fleet." 

This Japanese fleet, divided for the 
complicated plan into five major forces, 
with some of these split into smaller 
groups, steamed eastward independently 
to carry out the various phases of the sec- 
ond step in this strategy for 1942. Planes 
from two light carriers in the Second Mo- 
tile Force would strike Dutch Harbor, 
Alaska on 3 June to confuse the U. S. com- 

' Although some of the battleships knocked 
out at Pearl Harbor had been put back in serv- 
ice, and three others had been brought around 
from the Atlantic, these ships were oi>erating 
from the west coast as a final defense for the 

* In view of subsequent developments, Morison 
describes Yamamoto's disposition as "cockeyed." 
Coral Sea and Midway, 77-79. See also Battle 
That Doomed Japan, 73-78. These Japanese 
authors, although sometimes carefully kind to 
Yamamoto (who was killed later in a Solomon 
Islands air action) , are most often highly critical 
of the Japanese Navy and of war plans in general. 
Although the work is valuable and serious («. 6. 
the final two paragraphs of the book, at pages 
247-248), the authors sometimes sound like men 
on the morning after, ruefully surveying the night 




mand and to cover diversionary Japanese 
landings in the western Aleutians by the 
Occupation Forces for Adak-Attu and 
Kiska. Next the Carner Striking Force^ 
commanded by Vice Admiral Chuichi 
Nagumo, would soften Midway with the 
planes from the big fleet carriers Akagi, 
Kaga, Hiryti, and Soryu^^ and would then 
move on to strike the first blow at the U. S. 
Pacific Fleet if it challenged in a sortie 
from Pearl Harbor. 

Admiral Yamamoto's Main Body^ in- 
cluding three battleships and a light 
cruiser of his force plus the Aleutian 
Screening Force of four battleships and 
two light cruisers, then would go in for 
the kill against the U. S. Fleet. This en- 
gagement would be. followed, after dark- 
ness on 5 June, by Vice Admiral Nobutake 
Kondo moving in to shell the U. S. base 
for two days. Then Kondo's convoyed 
Transfort Group would approach to land 
the Midway Occupation Force of 5,000 
ground troops. While crossing the Pacific, 
Yamamoto remained some hundred?- of 
miles to the rear with his Main Boe&y^ 
awaiting word from the Advance Expe- 
ditionary Force of large fleet submarines 
already manning stations on the ap- 
proaches to Pearl Harbor to warn abotit 
sorties of the American ships. 

This ambitious plan might have worked, 
even though it was over- intricate. But 
again the Japanese had allowed their opti- 

mism and overconfidence to cast the U. S. 
Pacific Fleet in the role of a timid char- 
acter actor cued for a vulnerable "walk-on" 
part. They begged the question of tactics 
before their plan moved to the operational 
stage. The U. S. Fleet, according to Japa- 
nese plans, would be steaming for the 
Second Mohile Force in the Aleutians, or 
would be vacillating in Hawaiian waters, 
until the strong Carrier Striking Force hit 
Midway and revealed the target of the 
main effort.^ In either event nothing but 
the small Marine garrison force would 
stand in the way of the occupation of Mid- 
way, and the Japanese would have an air 
base of their own there before the U. S. 
Fleet could reach them. 

But, as at the Coral Sea encounter, the 
U. S. Fleet already had sortied to await the 
Japanese. For more than a month Nimitz 
had been aware that something like this 
was in the wind, and he bet nearly every- 
thing he had that the strike would hit Mid- 
way. The weakened Pacific Fleet stood 
some 300 miles northeast of Midway, there 
to refuel, before the Japanese picket sub- 
marines took position. As a result, these 
boats sighted no U. S. ships and radioed 
no reports, and Admiral Nagumo discov- 
ered the presence of the U. S. carriers in a 
most unpleasant manner. 

° Absent members of the original Pearl Harbor 
striking force were the Shokaku and the Zuikaku, 
the former undergoing Coral Sea damage repairs, 
and the latter reforming its air groups battered 
in that same action. Presence of these big car- 
riers at Midway might well have been decisive. 

^ Fuchida and Okumiya state that Japanese 
plans "calculated that the enemy naval forces 
would be lured out by the strike on Midway Island 
and not before." Battle That Doomed Japan, 128. 
But in this calculation their overconfidence must 
have been tempered somewhat, else why the di- 
version into the Aleutians? Morison discusses 
this faulty Japanese strategy, probably more real- 
istically. Coral Sea and Midway, 74-79. 


Midway Girds for Battle ^ 

Even before these Japanese plans were 
made, and long before Admiral Yamamoto 
sortied eastward, all U. S. military plan- 
ners recognized the vulnerable position of 
the Midway Atoll.^ Especially was this 
position clear in the light of early Jap- 
anese successes elsewhere in the Pacific, 
and none was more keenly aware of the 
grim situation than the atoll's small gar- 
rison force. The 12 PB Y's of VP-21 were 
soon withdrawn, and little help was ex- 
pected from the crippled fleet. But on 17 
December, while the 6th Marine Defense 
Battalion worked to improve existing de- 
fense installations, 17 SB2U-3's (Vindi- 
cators) of Marine Scout-Bomber Squad- 
ron 231 ( VMSB-231 ) flew in unexpectedly 
from Hawaii. Led by Major Clarence J. 
Chappell, Jr., and assisted in over- water 
navigation by a PB Y, the obsolescent craft 
made the 1,137-mile hop in nine hours and 
twenty minutes.^ Other reinforcements, 

^Editor's Note: Material contained in Chap- 
ters 3 and 4 is derived mainly from Chapters III 
and IV of the historical monograph Marines at 
Midway by Lieutenant Colonel Robert D. Heinl, 
Jr., published by Headquarters, U. S. Marine 
Corps, in 1948. This has been extensively re- 
written and checked against sources subsequently 
brought to light. 

' See Part II for a description of the geography 
and history of Midway. 

^ This was then the longest massed flight of 
single engine landplanes on record, and it had 
been carried out with no surface rescue craft 
available. CO MAG-21 serial 1173 to MGC, 
19Dec41. The flight took off from Hickam since 
Ewa's runways were too short for the heavily- 
laden planes to use with complete safety. 

including about 100 officers and men of 
Batteries A and C of the 4th Defense Bat- 
talion, left Pearl Harbor on 19 December 
with the old Navy 7-inch * and the 3-inch 
guns which had been shipped to Pearl Har- 
bor for Midway prior to the outset of war. 
(See Map 10, Map Section) 

This force, on board the XJSS Wright^ 
arrived on Christmas Eve, and Lieutenant 
Colonel Harold D. Shannon, who com- 
manded Marine defense forces on the atoll, 
turned over to Battery A (Captain Custis 
Burton, Jr. ) the mission of installing and 
manning the 7-inch and 3-inch batteries 
to be emplaced on Eastern Island. Battery 
C (First Lieutenant Lewis A. Jones) was 
assigned the job of setting up its 3-inch 
battory on Sand Island.^ 

Wext day Midway received another 
Christmas present: 14 Brewster F2A-3's, 
the air echelon of Marine Fighter Squad- 
ron 221 ( VMF-221 ) , flew in from the USS 
Saratoga which was retiring from the 
abortive attempt to relieve Wake Island. 
This squadron immediately began a daily 
schedule of air search and patrolling. On 
26 December the USS Tangier brought in 

* These 7-inch weapons had been removed from 
pre-World War I battleships and stored in reserve 
at naval yards. K. J. Bauer, "Ships of the Navy, 
1775-1945," (MS available from the author). 

^ Interview with LtCol C. Burton, Jr., 26Sep47, 
hereinafter cited as Burton. The two Eastern 
Island batteries were located side by side on the 
south shore of the island, near the western tip, 
and the Sand Island battery was set up along the 
north shore of Sand Island. 




Battery B of the 4th Defense Battalion 
(First Lieutenant Frank G. Umstead) ; 
additional machine gunners and 12 anti- 
aircraft machine guns from the Special 
Weapons Group of that same battalion; 
an aviation contingent of three officers and 
110 enlisted Marines constituting the 
ground echelon of VMF-221; aviation 
supplies; additional radar; and much- 
needed base-defense artillery material. 
Umstead's 5-inch battery, along with the 
island's other 7-inch battery, were set up 
south of the radio station on Sand Island. 
By New Year's Day of 1942 Midway was 
garrisoned by a strongly-reinforced de- 
fense battalion, and one fighter and one 
scout-bomber squadron. 

A major air base took shape on Eastern 
Island under the direction of Lieutenant 
Colonel William J. Wallace who on 9 
January became commanding officer of the 
Marine Aviation Detachment. Individual 
aircraft bunkers and underground person- 
nel shelters were built, emergency fueling 
expedients devised, radars calibrated, and 
inexperienced operators trained to use 
them properly. Colonel Wallace was as- 
sisted by Major Walter L. J. Bayler, the 
Marine aviation officer who had been sent 
back from Wake with that atoll's last 

The first test of this defense came on 
25 January during twilight general quar- 
ters when a Japanese submarine surfaced 
abruptly and opened fire on Sand Island, 
apparently trying to knock out the radio 
station. Battery D opened up with its 
3-inch guns and forced the enemy craft to 
crash-dive three minutes after it had sur- 
faced. Sand Island and the adjacent 
lagoon had received from 10 to 15 indis- 

• LtCol W. J. Wallace Itr to Col C. A. Larkin, 

criminate hits, and Captain Buckner's 
Battery D had expended 24 rounds. 

The next action against the atoll came 
two weeks later, on 8 February, when an- 
other submarine appeared less than 1,000 
yards south of Sand Island and opened 
fire on the radio towers. Captain Loren 
S. Fraser's Battery A opened fire on this 
boat, and it submerged after Marines had 
returned two rounds. The enemy had hit 
a concrete ammunition magazine, but for- 
tunately the small arms ammunition was 
not detonated. Two days later another 
submarine — or the same one — surfaced 
almost directly below two Marine fighter 
planes flying the sunset antisubmarine 
patrol. The submarine got off two rounds 
before First Lieutenant John F. Carey 
and Second Lieutenant Philip R. White, 
the fliers, could launch a diving attack. 
Both rounds from the submarine splashed 
in the lagoon, and then the boat was 
driven under water by the air attack just 
as the batteries of the 6th Defense Bat- 
talion were going into action. This was 
the last time for a number of months that 
Midway was troubled by enemy sub- 

As the winter wore on, Midway's air arm 
began to profit from the general expansion 
of Marine Corps aviation, and the two 
squadrons and their small provisional 
headquarters on 1 March became Marine 
Aircraft Group 22 (MAG-22). On 20 
April Lieutenant Colonel Wallace was suc- 
ceeded in command by Major Ira L. Kimes, 
and at the same time Major Lofton R. 
Henderson took command of VMSB-241, 
(the new designation of former VMSB- 
231 ) . This was a busy time for MAG-22, 
which was then converting Eastern Island 
from a small advanced air base to a major 
installation capable of handling as many 
squadrons and aircraft types as could 



physically be accommodated and protected 

On 10 March, during the work and re- 
organization, the Marine fliers got their 
first test against enemy aircraft. Radar 
picked up a Japanese four-engined 
"Mayis" (Kawanishi 97) approximately 
45 miles west of Midway, and 12 fighters 
under Captain Robert M. Haynes vectored 
out to intercept. They made contact with 
the enemy flying boat at 10,000 feet and 
shot it down. 

Although the enemy plane did as well as 
it could to fight off this attack, this contact 
was more important as intelligence for 
Nimitz' staff in Pearl Harbor than as a test 
for the Marine fliers. Two aircraft of this 
type had tumbled four bombs into the hills 
behind Honolulu on the night of 3-4 
March, and Nimitz already believed that 
this portended an offensive toward Hawaii. 
Now this new sighting near Midway gave 
added weight to his estimate, and this in- 
formation went into the CinCPac intelli- 
gence "hopper" which shortly thereafter 
reached the considered opinion that the 
Japanese attack would strike Midway. 
By this time the Japanese code had been 
broken, also.' Thus were the fragments 
pieced together into Nimitz's May 1942 
decision which caused him to wager nearly 
every ship he had in an early sortie from 
Pearl Harbor to the position 300 miles 
northeast of Midway from which the 
Japanese could be intercepted. 

It was a bold, even though well- 
calculated, wager. The many ships on 
South Pacific convoy duty had to be left 
on their important jobs; Halsey's Enter- 
prise and Hornet had rushed from the 
Doolittle launching area part way to the 

Carol Sea and back again to Pearl Harbor 
where they were placed under a new com- 
mander. Rear Admiral Raymond A. 
Spruance; Fletcher's Yorktown had just 
limped in from the Coral Sea needing an 
estimated 90 days of repair work ; and all 
the Fleet's battleships were on the west 
coast where they could not be used (partly 
because Nimitz felt he did not have 
enough air strength to protect them, any- 
way) . But Nimitz was convinced that his 
intelligence estimate was correct, and that 
the stand had to be made. 

For the engagement Nimitz gave 
Fletcher, in over-all tactical charge,* direct 
command of Task Force 17 which included 
the Yorktown (rushed into shape in two 
days rather than 90) , two cruisers, and six 
destroyers. Spruance commanded Task 
Force 16 which included the Enterprise 
and Hornet, six cruisers, and nine destroy- 
ers. Four oilers and 19 submarines also 
were assigned to the area, and, in addition, 
a North Pacific Force was formed of five 
cruisers and ten destroyers to screen the 
Aleutians. The Japanese had him out- 
numbered on all counts, and Nimitz knew 
that the enemy would be gunning for the 
three U. S. carriers. But his carriers like- 
wise were his only hope, and the admiral 
ordered his subordinates to apply the rule 
of calculated risk when they went in with 
their air groups to stop the Japanese. 

While Nimitz readied this reception 
committee, the Japanese completed their 
Midway plans and polished the rough op- 
erational edges with carrier training and 

''Battle That Doomed Japan, 131; C/.,<S. A Sea 
Power, 686. 

' Fletcher was senior to Spruance, and thus be- 
came OflScer in Tactical Command. But as it 
turned out, Spruance exercised practically an in- 
dependent command during the critical days of 
4-6 June, and this probably was fortunate be- 
cause Fletcher had no aviation staff while Spru- 
ance had inherited Halsey's. 



rehearsals. By the last week of May, all 
Japanese fleet elements were underway, 
and on decks Imperial sailors sunbathed 
and sang songs — vocal eruptions of what 
has been described as the "Victory 
Disease." *• 

Meanwhile on Midway, the focal point 
for these vast efforts, Marines got their 
first inkling of all this attention when 
Nimitz flew in on 2 May to see their senior 
officer. Lieutenant Colonel Shannon, and 
the atoll commander. Commander Cyril 
T. Simard. The admiral inspected the in- 
stallations, and then directed Shannon to 
submit a detailed list of all supplies and 
equipment he would need to defend the 
atoll against a strong attack. Nimitz 
promised that all available items requested 
would be forwarded immediately, and 
within less than a week men and material 
were being embarked in Hawaii to bolster 
the island strength. 

Three more 3-inch antiaircraft batteries 
totaling 12 guns, a 37mm antiaircraft bat- 
tery of eight guns, and a 20mm antiaircraft 
battery of 18 guns were temporarily de- 
tached from the 3d Defense Battalion at 
Pearl Harbor ; and two rifle companies of 
the 2d Marine Raider Battalion, together 
with a platoon of five light tanks, also were 
sent along to Midway. For MAG-22, still 
flying Brewster fighters and Vought Vin- 
dicator dive bombers, there would be 16 
SBD-2 dive bombers and seven relatively 
new Grumman F4F-3 fighters. 

Shortly after his return from Midway 
to Pearl Harbor, Nimitz arranged "spot" 
promotions to captain and colonel respec- 
tively for Simard and Shannon, and de- 
scribed to them in a joint personal letter 

the steps being taken to reinforce their 
atoll against the anticipated attack. 
Japanese D-Day, the admiral predicted, 
would be about 28 May. On the day they 
received this letter, Simard and Shannon 
conferred on their final plans for defense, 
and that evening Colonel Shannon as- 
sembled his key subordinates and warned 
them of the impending enemy attack. 
Additional defensive measures and priori- 
ties of final efforts were outlined, and all 
recreational activities suspended. May 25 
was set as the deadline for completion of 
the measures ordered. 

On the 25th, however, came two wel- 
comed changes for the picture. First, 
Nimitz passed the word that the Japanese 
attack was not expected until early June, 
and, second, the first reinforcements ar- 
rived. On this date the USS St. Louis 
came in with the 37mm antiaircraft bat- 
tery of the 3d Defense Battalion plus the 
two companies of raiders. Four of the 
37's were emplaced on each island while 
Raider Company C (Captain Donald H. 
Hastie) went to Sand Island, and Com- 
pany D (First Lieutenant John Apergis) 
to Eastern Island. 

Next day the aircraft tender Kittyhawk 
arrived with the 3d Defense Battalion's 
3-inch antiaircraft group commanded by 
Major Chandler W. Johnson, the light 
tank platoon for the mobile reserve, and 
the SBD-2's and the F4F-3's. In the 
following week additional Army and 
Navy planes arrived, and by 31 May there 
were 107 aircraft on the island.^" 

"Battle That Doomed Japan, 245. ". . . the 
spread of the virus was so great," the authors say, 
"that its effect may be found on every level of the 
planning and execution of the Midway operation." 

" By this date the daily aviation gasoline con- 
sumption of planes based on Eastern Island was 
65,000 gallons, and the following numbers of 
planes were based there : U. S. Army — four 
B-26's and 17 B-17's ; U. S. Navy— 16 PBY-5A's 
and six TBF's ; U. S. Marine Corps— 19 SBD-2's, 
17 SB2IT-3's, 21 F2A-3's, and seven F4F-3's. 



For the ground forces and key civilian 
workers who had remained behind to help 
defend the island, the week was equally 
busy. Reinforcing weapons were in- 
stalled, tanks tested in the sand, all defen- 
sive concentrations registered in, and the 
emplacing of an extensive system of ob- 
stacles, mines and demolitions completed. 
Sand Island now was surrounded with two 
double-apron barbed wire barriers, and all 
installations on both islands were ringed 
by protective wire. Antiboat mines of 
sealed sewer pipe, and obstacles of rein- 
forcing steel lay oflFshore ; the beaches were 
sown with homemade mines of ammuni- 
tion boxes filled with dynamite and 20- 
penny nails ; cigar box antitank mines cov- 

ered likely beach exits; and bottles of 
molotov cocktail stood ready at every posi- 
tion. A decoy mockup airplane (a JFU — 
Japanese fouler-upper) was spotted prom- 
inently on the seaplane apron, and all un- 
derground fuel storage areas on Sand 
Island were prepared for emergency de- 
struction by demolition. ^^ 

" The demolition system worked, too. On 22 
May a sailor threw the wrong switch and blew 
up a good portion of the aviation gasoline. The 
supply was so critical after this that the pilots 
who arrived on the Kittyhawk did not get a 
prebattle chance to check out in their SBD-2's, 
Pipe lines also were wrecked in the blast, and 
MAG-22 thereafter had to refuel all planes by 
hand from 55-gallon drums. 


Midway Versus the Japanese 
4-5 June 1942 

A Midway PBY spotted the approach- 
ing Japanese first, at about 0900 on June 
3,^ and tracked them long enough to re- 
port eleven ships making 19 knots east- 
ward. These vessels were probably the 
transport and seaplane groups of the Occu- 
pation Force, and they were attacked at 
1624 by nine B-l7's which Captain Simard 
sent out following the PBY's contact re- 
port. The pilots reported having hit "two 
battleships or heavy cruisers" and two 
transports in the group then 570 miles 
away from Midway, but the fliers were 
mistaken in both ship identification and 
in calling their shots, for they actually 
hit nothing. A Catalina scored on one of 
these oilers later that night in a moon- 
light torpedo run. 

This was enough to convince Fletcher 
that the battle would soon be on, and he 
changed course from his station 300 miles 
east-northeast of Midway to gain a new 
position about 200 miles north. From 
there he could launch his planes the fol- 
lowing morning against the Japanese car- 
rier force which was expected to come in 
from the northwest. U. S. intelligence 
still was good. Nagumo continued to steam 
in from the northwest while his transports 
were under attack, and near daybreak on 
4 June, when the Yorktawn launched an 
early -morning search and while 11 PBY's 

' Midway (Zone plus 12) time and West Longi- 
tude date. 

were going up to patrol from Midway, he 
had reached a position approximately 250 
miles northwest of the atoll. There at 
0430 the Japanese admiral launched 36 
"Kate" torpedo planes and 36 "Val" dive 
bombers, plus 36 escorting Zeros, for the 
first strike against the atoll. 

At 0545 one of the Midway PBY's 
sighted these planes about 150 miles out 
from the island, and a short time later 
another PBY reported visual contact with 
two enemy carriers and the balance of the 
Japanese Carrier Striking Force some 200 
miles from Midway. Enterprise inter- 
cepted this report, but Fletcher wanted to 
recover his search planes and sift further 
intelligence before launching his strike, 
and so he ordered Spruance to take the van 
southwesterly and lead off the attack 
against the enemy carriers. 

Meanwhile the Midway Marines were 
ready for the first shock of attack. Ground 
force defenders at general quarters manned 
every weapon and warning device, and 
MAG-22, which already had fighters up to 
cover the sortie of the PBY's stood by for 
orders. At 0555, shortly after the second 
PBY report had fixed the position of the 
Japanese Striking Force, the 6th Defense 
Battalion radar logged a report of "many 
planes," and the Naval Air Station raised 
similar blips almost simultaneously. Air 
raid sirens began to wail. Condition One 
was set, and the MAG-22 pilots manned 
their planes. Both squadrons were in the 




CAMOUFLAGED LOOKOUT TOWER AT SAND ISLAND stands amidst the damage caused by 
Japanese dive bombers which attacked Midway Atoll on 4 June 1942. (USN 17057 ) 

JAPANESE CRUISER MIKUMA, sunk at the Bailie uj Midway, lies balleted and smoking from the 
attacks of pilots of MAG-22 and the American carriers. (USN 11528) 



air in less than 10 minutes, VMF-221 
heading to intercept the enemy planes and 
VMSB-241 off to rendezvous station 20 
miles east where the dive bomber pilots 
would receive further instructions. 

The VMF fliers under Major Floyd B. 
Parks sighted the Zero-escorted Val dive 
bombers at 0616 about 30 miles out from 
Midway, and Captain John F. Carey, 
leading one of Parks' divisions in an 
F4F-3, launched the attack from 17,000 
feet. The Marine fliers were hopelessly 
outnumbered, and they found that the 
Zero fighters could "fly rings around 
them." They had time for only one pass 
at the bombers, and then had to turn their 
attention to the swarm of Zeros, from one 
to five of which got on the tail of each 
Marine fighter. Only three of the original 
12 Marine pilots survived this brawl, and 
although the damage they inflicted on the 
enemy has never been assessed, it is be- 
lieved that they splashed a number of the 
bombers and some of the Zeros. Other 
Zeros were led into the Midway antiair- 
craft fire. 

Meanwhile another group of 13 Midway 
fighters under Captain Kirk Armistead 
came in for an attack against the enemy 
air formation. Again the damage inflicted 
upon the enemy was undetermined, but 
fewer Marine pilots were lost. For better 
or for worse, however, the fighter defense 
of Midway had been expended, and the 
problem now passed to the antiaircraft 
guns on the atoll. 

The first Japanese formation attacked 
at about 0630 from 14,000 feet. Antiair- 
craft fire knocked down two of these hori- 
zontal bombers before they could unload, 
but 22 came on through to drop their 
bombs. And just as these initial explo- 
sions rocked the two islands, 18 planes of 
the enemy's second wave came over for 

their strike. Since each of these Japanese 
formations had left the carriers with 36 
planes, it is possible that the Marine fliers 
scored some kills.^ 

The Kaga aircraft group in the first 
wave, assigned to attack the patrol plane 
facilities on Sand Island, dropped nine 
242-kilogram bombs on and about the sea- 
plane hangars, setting them aflame and 
starting a large fire in the fuel oil tanks 
500 yards to the north. The Akagi planes 
plastered the north shore of Eastern 
Island to destroy the Marine mess hall, 
galley, and post exchange. These the re- 
turning enemy fliers described as hangars.' 
Other targets of the Japanese dive 
bombers included the already-flaming fuel 
storage at the north end of Sand Island, 
the Sand Island dispensary, and the East- 
ern Island powerhouse which suffered 
direct hits from two 805-kilogram bombs. 
These hits virtually destroyed the entire 
plant. And at the very end of the strike, 
the 6th Defense Battalion's Eastern Island 
command post received a direct hit which 
killed the Marine sector commander, 
Major William W. Benson, and wounded 
several other men. After these bombers 
completed their runs, the Zeros came in 
for strafing attacks. This one and only 
air strike on Midway was over shortly 
after 0700. 

^ Maj W. S. McCormick, an experienced anti- 
aircraft oflBcer, counted the 22 ; and Capt M. A. 
Tyler, a VMSB-241 pilot with a grounded plane, 
counted the 18. CO, VMF-221 Kept to CO, MAG- 
22, 6Jun42, 1. However, Battle That Doomed 
Japan, 155, reports that "not a single hit" was 
sustained by the Japanese bombers until they 
struck in two waves of 36 planes each. This 
seems highly improbable in view of eyewitness 
accounts and damage sustained. 

' ONI Review, 45-48. Information on ground 
defense from CO, 6th DefBn Rept to CO, NAS, 
Midway on action of 4-5Jun42, 13Jun42, 1-8. 



Marine defense batteries fired through- 
out these attacks, and one source credited 
this antiaircraft fire with 10 kills.* Ee- 
ports from Marine flyers would appear to 
require an increase of this estimate, how- 
ever, since returning Midway pilots de- 
scribed enemy planes falling out of 
formation and others floundering into the 
water.^ But Japanese authorities claim 
that only six of their planes — three level 
bombers, two dive bombers, and a fighter — 
failed to make it back to the carriers.^ 
This controversy probably will never be re- 
solved, for regardless of how many of these 
Japanese planes made it back to their car- 
rier decks, Fletcher and Spruance — with a 
certain unintentional assistance from 
Nagumo — initiated action which resulted 
in the destruction of all these planes, 

Nagumo's mistake was a natural one for 
a commander who believed himself to be 
unopposed on a "field" of battle of his own 
choice. Lieutenant Joichi Tomonaga, the 
flight officer who had commanded the first 
attack wave against Midway, radioed dur- 
ing his return flight that "There is need for 
a second attack wave." Meanwhile, with 
Nagumo still ignorant of the U. S. fleet's 
presence in the vicinity, six American 
TBF's and four B-26's from Midway came 
in to attack his ships. This convinced the 
Japanese admiral that Tomonaga was 
right, and he sent below to hangar spaces 
the 93 planes he had kept spotted for 
strikes against possible surface opposition. 
These planes were to be re-armed with 
bombs for the second strike. Then 
Nagumo called in the returning planes to 

arm them for the new attack of the atoll. 
While his men were involved in this work 
on the flight deck and in hangar spaces, 
Nagumo got the belated word from a Tone 
search plane that U. S. ships, including at 
least one carrier, were in the area. This 
caused another change of mind, and the 
admiral ordered the planes' ordnance 
changed again, from bombs back to tor- 
pedoes with which to attack the surface 
ships. But this decision was just tardy 
enough to allow Spruance to catch him 
with his planes down, and with torpedoes 
and bombs strewn in great confusion about 
the hangar deck.^ 

Meanwhile, as Nagumo vacillated, Ad- 
miral Nimitz's orders for Captain Simard 
to "go all out for the carriers," while 
Marine antiaircraft batteries worried 
about Midway, were under execution. 
VMSB-241, like the fighter squadron, had 
divided into two striking units, the first 
composed of 16 SBD-2's led by Major 
Lofton Henderson, and the second of 11 
SB2U-3's commanded by Major Benjamin 
W. Norris. Henderson's group climbed to 
9,000 feet to locate the enemy carriers, 
which were then undergoing the attack 
from the TBF's and the B-26's. Fliers of 
this group sighted the Japanese ships at 
0744, but as the SBD's spiralled down they 
were set upon by swarms of Nakajima 97's 
and Zeros flying air cover, which were 
soon reinforced by more fighters from the 
carriers below. Henderson and several 
others were shot down (only eight of these 
planes got back to Midway) and the strike 
scored no hits although some were 

* ONI Review, 72. 

"CinCPac Rept to CominCh on the Battle of 
Midway, 8. 
'Battle That Doomed Japan, 156. 

' ONI Review, 17-19, 44-^5. 

* Statement of Capt E. G. Glidden, 7Jun42, 1. 
Japanese sources disclose, however, that no hits 
were scored in this attack. The Guadalcanal 



Next came an attack by 15 B-l7's led by 
Lieutenant Colonel Walter C. Sweeney, 
USA, but again claims of hits were opti- 
mistic. And as these Flying Fortresses 
pulled away, Major Norris came in with 
his 11 Vindicators which had taken off 
with Henderson. Beset by the Zeros, 
Norris turned to the nearest target at hand, 
and the Marines crowded their ancient 
planes into a standard glide run almost 
on top of the Japanese battleship 
Haruna — previously claimed as an Army 
B-17's victim off Luzon. Some of the 
fliers also went after the Kirishima, which 
was nearby, but neither attack managed 
any hits. Three Marines were shot down, 
and the group was credited with splashing 
two enemy fighters, plus two probables.® 

By 1100 all surviving Marine aircraft 
had made their way back to the atoll 
where all hands grimly assessed the battle's 
damage and prepared for subsequent ac- 
tion. Of the VMF-221 fighters which had 

airfield, captured two monthis later, was named 
in Maj Henderson's honor. Rear gunners of this 
strike group are credited with four enemy kills 
plus two additional probables. 

° ONI Review, 19 ; VSSBS Interrogation^^ Nav 
No 2, Capt Susumu Kawaguchi, IJN,I,6. See 
also Coral Sea and Midway, 111, for Adm Mor- 
ison's dismissal of damage claims by land-based 

gone in against the attacking Japanese 
planes, only 10 returned, and of this num- 
ber only two were in shape to leave the 
ground again. Thirteen F2A-3's and two 
F4F's were missing, along with the eight 
craft lost from the Henderson group and 
the three shot away from the Norris force. 
Slick black smoke from oil fires billowed 
up from the islands, and ruptured fuel 
lines left more than two-thirds of the avia- 
tion fuel temporarily unavailable. Gaso- 
line had to be sent to the field from Sand 
Island, and hand-pumped from drums. 
The Marine ground defense force had sus- 
tained 24 casualties, and four ordnance- 
men of VMF-221 had been lost to a direct 
bomb hit. 

At 1700 a burning enemy carrier was re- 
ported 200 miles northwest of Midway, 
and Major Norris prepared VMSB-241's 
six operational SBD-2's and five SB2U-3's 
for a night attack. The planes took off 
at 1900, but could not find the carrier. 
Major Norris failed to return from this 
mission, although the other pilots managed 
to home by the light of oil fires and 
the antiaircraft searchlights which were 
turned up as beacons.^" Meanwhile, the 
Battle of Midway had been decided at sea 
in a fiofht of carrier aircraft. 

' VMSB-241 Rept of Combat, 7Jun42, 3. 


Battle of the Carrier Planes 
4 June 1942 

While the land-based fliers had their 
morning go at the Japanese Striking 
Force, and while Nagumo juggled his 
planes and decisions, Spruance steamed 
southeast to lead off the attack against the 
enemy. The American admiral intended 
to hold his planes until he drew within 
about 100 miles of the Japanese. But when 
he heard of the strike on Midway, Spru- 
ance- launched two hours before this in- 
tended range would have been reached. 
By this calculated risk he hoped to catch 
the Japanese planes back on their carriers 
rearming for a second attack of the atoll. 
And about twenty minutes later Nagumo 
made the decision which set up himself 
and his planes as exactly the target Spru- 
ance hoped his pilots would find.^ 

Enterprise and Hornet began launching 
at about 0700, sending off every operational 
plane they carried, except a few to cover 

' In his introduction to Battle That Doomed 
Japan, Spruance writes : "In reading the account 
of what happened on 4 June, I am more than ever 
impressed with the part that good or bad for- 
tune sometimes plays in tactical engagements. 
The authors give us credit, where no credit is 
due, for being able to choose the exact time for 
our attack on the Japanese carriers when they 
were at the greatest disadvantage — flight decks 
full of aircraft fueled, armed and ready to go. 
All that I can claim credit for, myself, is a very 
keen sense of the urgent need for surprise and a 
strong desire to hit the enemy carriers with 
our full strength as early as we could reach 

the task force. The strike was led by 29 
Devastator (TBD-1) torpedo bombers, 
and these were followed by 67 Dauntless 
dive bombers and 20 Wildcat fighters. 
Eighteen other Wildcats, plus a like num- 
ber withheld to relieve them later, pa- 
trolled overhead. Yorktown held back its 
planes for about two hours ; Fletcher con- 
sidered that the aircraft from his carrier 
might be needed against other enemy car- 
riers not yet located, but by 0838 there had 
been no enemy sightings, and he decided 
to launch half his dive bombers and all 
his torpedo planes, along with escorting 
fighters. By shortly after 0900 the York- 
town liad 17 SBD's, 12 TBD's, and six 
F4F-3's in the air, and other planes ready 
for takeoff. 

As Spruance had hoped, Nagumo con- 
tinued for more than an hour to steam to- 
ward Midway, and the first U. S. planes 
found the Japanese Carrier Sticking Force 
with its flattops in the center of a larger 
formation consisting of two battleships, 
three cruisers, and 11 destroyers. By 
0917, Nagumo had recovered his Midway 
attack planes, and at that time he made a 
90-degree change of course to east-north- 
east. This course change caused 35 of the 
Hornefs SBD's and escorting fighters to 
miss the battle, but Hornefs torpedo 
planes found the enemy and went in low 
without fighter cover. 

The 15 obsolete Devastators met heavy 
antiaircraft fire from the Japanese Strik- 




ing Force, and pulled down upon them- 
selves the bulk of the Zeroes patrolling 
overhead. Against this combined fire, few 
of the planes got close enough to Japanese 
ships to launch torpedoes, and again, as in 
the Coral Sea battle, any hits scored by the 
slow unreliable torpedoes of that period 
proved duds. This antiaircraft and 
fighter opposition started while the planes 
had yet eight miles to go to reach the Japa- 
nese ships, and only one Devastator pilot 
lived to pull up from this attack.^ 

The 14 TBD's from the Enterprise fared 
only a little better. Four of these planes, 
likewise striking without fighter escort, 
survived their torpedo runs against the 
Japanese ships, although they scored no 
hits. But these two Devastator attacks, 
costly as they were, served to pull down the 
Zero canopy to such a low altitude that tlie 
following SBD's from Enterprise and 
Yorktoum had an easier time of it. 

These Dauntless dive bombers came in at 
about 1020 while Nagumo's ships still were 
dodging the Devastators. The Akagi took 
two hits which set her afire, and Admiral 
Nagumo transferred his flag to the light 
cruiser Nagara? The Kaga sustained 
four hits, and at 1925 she blew up and sank. 
The Soryu was hit three times by the planes 
and also struck by three torpedoes from the 
submarine Nautilvs which arrived on the 
scene between 1359 and 1495.* Finally the 

' Ensign George H. Gay was this sole survivor. 
His plane splashed shortly after he had pulled 
up from his run which had skimmed a carrier 
deck at about 10 feet. Gay's gunner had been 
killed, but Gay was rescued from his life raft next 
day by a Catalina. 

' Although the Akagi remained afloat, she was 
abandoned at 1915 and later scuttled by a torpedo 
from one of her screening destroyers. 

^Battle That Doomed Japan, 185, 189-191, 
presents strong evidence to indicate that this was 

Soryu's gasoline stowage exploded and 
broke the ship in half. 

By 1030, Nagumo had lost the services of 
three carriers, and in all three cases, as 
Spruance had hoped, the American attack 
had caught the ships in process of refuel- 
ing and rearming the planes of their Mid- 
way strike. But even with these ships and 
their planes gone, Nagumo still was deter- 
mined to fight back with his surviving car- 
rier, the Fliryu, which had escaped damage 
by getting far out of position in some of 
the earlier evasive actions to escape the tor- 
pedo planes. "Although defeat now stared 
the Japanese starkly in the face, they felt 
that the battle had to be continued as long 
as we retained even a small part of our 
striking power." ^ 

When the Akagi was shot from under 
Nagumo, the Japanese Striking Force com- 
mander temporarily passed his command 
to Rear Admiral Hiroaki Abe on board the 
heavy cruiser Tone, and command of air 
operations simultaneously passed to Rear 
Admiral Tamon Yamaguchi in the Hiryu. 
At about 1050 two float planes from the 
cruiser Chikuma sighted the Yorktown 
iask group and guided to it a strike of 18 
dive bombers and six fighters up from the 
Hiryu. The U. S. air patrol and antiair- 
craft fire knocked down or turned back 
most of these enemy planes which arrived 

not the Soryu but the Kaga, and that the one tor- 
pedo which actually hit proved a dud. 

^Battle That Doomed Japan, 191. In Coral 
Sea and Midway, at page 132, Adni Morison 
points out that the Japanese at this time assumed 
from scout plane reports that the U. S. force had 
no more than two carriers, and possibly only one. 
The Japanese authors in Battle That Doomed 
Japan point out on page 174, however, that while 
the torpedo planes were yet approaching for their 
first strike against Nagumo, "Reports of enemy 
planes increased until it was quite evident that 
they were not from a single carrier." 



at about noon, but those that got through 
scored three hits which started fires. 
Within 20 minutes tlie big carrier was dead 
in the water. 

Her crew got her underway again in 
about an hour, but a second strike from the 
Hiryu appeared early in the afternoon. 
Although the Japanese lost half of the 10 
Kate torpedo bombers and six Zero fight- 
ers of this attack, four of the Kates came in 
to attack the Yorktown at masthead level. 
Launching at a range of about 500 yards, 
two of the torpedo planes scored hits which 
left the carrier not only dead in the water 
but listing so badly that she was abandoned 
a few minutes later ."^ 

Meanwhile, one of YorhtowrCs search 
planes (one of 10 scout bombers sent out 
before the first attack on the ship ) spotted 
the Hiryu., two battleships, three cruisers, 
and four destroyers at 1445, and reported 
the position of these enemy ships. At 1530 
Spruance had 24 SBD's ^ up from the En- 
terprise, and they found the Hiryu and her 
screening ships at 1700. Using the same 
tactics which had paid off in their morning 
attacks, the dive bombers scored four hits 
which finished operations for Nagumo's 

° The speed with which her crew had put York- 
toum. in shape after the first attack led the Japa- 
nese to believe that this second strike was against 
a different carrier. They had by now spotted all 
three U. S. carriers, but at this point they thought 
they had destroyed two of them. This second 
strike still did not finish the battered carrier, 
however. She remained afloat and regained 
some degree of equilibrium without human aid. 
Salvage parties went on board the following day, 
and the ship was taken under tow. But one of 
Nagumo's float planes spotted her early on 5 
June, and a submarine was sent out to finish her 
off. The sub found the carrier on the 6th, put 
two torpedoes in her, and she finally went down 
early on 7 June. 

'Ten of these were refugees from Yorktown, 
and the others veterans of the earlier strikes. 

fourth and last flattop.* The bombing 
cost three SBD's and their crews. 

During all this action Admiral Yama- 
moto, still miles to the rear, considered 
himself fortunate to have drawn out the 
U. S. Pacific Fleet. Shortly after noon, 
when he heard of the Hiryu's first strike 
against the Yorktown, the Japanese com- 
mander ordered the Aleutian Screening 
Group and Admiral Kondo's Second Fleet 
to join his MoAn Body by noon the next 
day to finish off the U. S. ships and com- 
plete the occupation of Midway. And a 
full hour and twenty minutes after he 
heard of the fate of Nagumo's final car- 
rier, Admiral Yamamoto sent out a mes- 
sage in which he reported the U. S. fleet 
"practically destroyed and . . . retiring 
eastward," and he called on Nagumo, the 
Invasion Force (less Cruiser Division 7) 
and the Suhm<irine Force to "immediately 
contact and attack the enemy." A stimu- 
lating message, but "In the light of the 
whole situation ... so strangely optimistic 
as to suggest that Commander in Chief was 
deliberately trying to prevent the morale 
of our forces from collapsing." ^ 

Nagumo's morale obviously needed to 
be stimulated by stronger stuff ; at 2130 he 
reported : "Total enemy strength is 5 car- 
riers, 6 heavy cruisers, and 15 destroyers. 
They are steaming westward. We are re- 
tiring to the northwest escorting Hiryu. 
Speed, 18 knots."" Yamamoto's answer 

" The Hiryu floated in flames until, as in the 
case of the Akagi, one of the ships of her screen- 
ing force put her to death with torpedoes at 0510 
next morning. ONI Revieic, 13. 

"Battle That Doomed Japan, 213. 

"° It)i<i. To which message the authors quote 
a response by one of Yamamoto's staff officers, 
who, they say, "voiced the dejection of the entire 
Combined Fleet staff. . . ." Made by Rear Ad- 
miral Matome Ugaki, Yamamoto's chief of staff, 



relieved Nagumo of command in favor of 
Rear Admiral Kondo; but later messages 
told the commander in chief that there was 
little chance of finding the U. S. Fleet until 
after dawn next day. At 0255 on 5 June 
the admiral changed his mind, abandoned 
the Midway venture, and ordered a general 

Admiral Spruance, now more on hi •. own 
than ever, following Fletcher's move from 
the damaged Yorktown to the Asforia,^'^ 
of course did not know of Yamamoto's 
decision ; but he did know that vastly more 
powerful enemy surface forces could well 
be nearby, quite possibly with additional 
carriers that had come in with the Main 
Body or with another enemy force. His 
problem, as he saw it, was to avoid combat 
in which he could be hopelessly outclassed, 
especially at night, and yet at the same 
time keep within air support distance of 
Midway in case the Japanese should per- 
sist in their assault plans. This he suc- 
ceeded in doing, but in the process lost 
contact with the enemy fleet. He did not 
regain contact until 6 June. 

In the early morning hours of 5 June, 
however, a retiring Japanese column of 
four cruisers and two destroyers was 
sighted by U. S. submarine Tainbor; and 
when the Japanese sighted the Tmnbor, 
evasive action resulted in a collision of 
their cruisers, Mogaini and Mikwma. 
While the other Japanese cruisers re- 
tired at full speed, these two lagged behind 

the statement must likewise be considered a 
classic of understatement : "The Nagumo Force 
has no stomach for a night engagement!" 

" Shortly after 1300 on 4 June, Spruance ra- 
dioed his disposition and course to Fletcher on 
board the Astoria, and asked if Fletcher had 
instructions for him. Fletcher replied : "None. 
Will conform to your movements." Coral Sea 
and Midway, 141n. 

with the destroyers to screen them, the 
Mogami with a damaged bow and the 
Mikuma trailing oil. The submarine con- 
tinued to stalk these four ships, did not 
manage to gain a firing position, but at 
break of day reported their position. 

Captain Simard sent 12 B-l7's out from 
Midway to attack these ships, but the 
Flying Fortresses had trouble locating 
their targets, and Simard then ordered a 
Marine bombing squadron off on this mis- 
sion. Captain M. A. Tyler with six 
SBD-2's and Captain Richard E. Fleming 
with six SB2U-3's took off at about 0700 
to attack the ships which were then re- 
ported to be 170 miles west of the atoll. 
They located the ships at about 0800, and 
Tyler led his division out of the sun to- 
ward the stern of the Mogami wliile Flem- 
ing and the other Vindicator pilots went 
down at the Mikuma. 

Both divisions met heavy antiaircraft 
fire, but Tyler and his fliers bracketed 
their target with six near misses which 
caused some topside damage to the Mo- 
gamAP Fleming's plane was hit, but the 
pilot stayed on course at the head of his 
attack formation and crashed his plane 
into the Mikuma! s after gun turret." This 
additional damage further slowed the 

" U88BS Interrogations, Nav No 83, RAdm 
Akira Soji, I, 363. Adm (then Capt) Soji had 
command of the Mogami during this action. 

" Fleming's dive ". . . crashed into the after 
turret, spreading fire over the air intake of the 
starboard engine room. This caused an explo- 
sion of gas fumes below, killing all hands work- 
ing in the engine room. This was a damaging 
blow to the cruiser, hitherto unscathed except 
for the slight hull damage received in the col- 
lision with Mogami. Both cruisers were now 
hurt, and they continued their westward with- 
drawal with darkening prospects of escaping 
the enemy's fury." Battle That Doomed Japan, 



cruisers, and Admiral Spruance's carrier 
planes found the cripples the following 
day, the 6th. The attack of these planes 
sank the Mikuma and inflicted enough 
additional damage on the Mogami to keep 
her out of the war for the next two years. 

The Battle of Midway — which many his- 
torians and military experts consider the 
decisive naval engagement of the Pacific 
War — was over, and all actions following 
those of 4 June were anti-climactic. The 
U. S. had lost 98 carrier planes of all types, 
and would lose the Y orktown, then under 
tow. The Japanese carriers sustained 
total losses of about 322 planes of all 
types.^^ And with the four carriers had 
gone the cream of their experienced naval 
pilots. This, along with later losses in air 
battles over Guadalcanal, was a blow from 
which the Japanese never fully recovered." 

Although the carrier planes had decided 
the large issue, the contribution of Marines 
to the defense of Midway had been consid- 
erable, from the inception of base develop- 
ment to the end of the action. Not only 
had the 3d and 6th Defense Battalions con- 
tributed their share of labor, vigilance, and 
flak, but the aviation personnel of MAG- 
22, at a cost rarely surpassed in the history 
of U. S. naval aviation, had faced a 
superior enemy and exacted serious dam- 
age. At a cost of 49 Marines killed and 53 
wounded, Midway had destroyed some 43 
enemy aircraft (25 dive bombers and 18 
Zeros) in air action, plus another 10 shot 
down by antiaircraft guns.^" 

"Ibid., 250. This figure may be suspect. It 
exceeds considerably the regular complement of 
the four carriers. 

^ For a discussion of the "crack-man policy" of 
the Japanese Navy Air Force, see Battle That 
Boomed Japan, 242-243. 

" These are reported figures based on the sound- 
est possible estimates. The Japanese account in 

In another air-air action, similar to that 
at Coral Sea, Fletcher and Spruance had 
sent the proud Imperial Fleet scurrying 
home to Japan without firing a shot from 
its superior naval rifles. Yamamoto could 
gain little consolation from the fact that 
the northern operation had secured two 
Aleutian bases : what good the bowl if the 
rice is gone? For". . . unlike most of the 
Nipponese war lords, [Yamamoto] ap- 
preciated American strength and re- 
sources." " He knew that destruction of 
the U. S. Fleet early in 1942 was a neces- 
sary prerequisite to the year's plans for 
control of the Coral Sea and the American 
sea lanes to Australia and New Zealand, 
and, in the final analysis, the necessary 
prerequisite to the success of Japan's en- 
tire war effort." 

But now that the Japanese clearly were 
defeated at Midway, they no longer could 
overlook the setback they had received at 
Coral Sea in phase one of their 1942 plans, 
and phase three — occupation of the Fijis, 
Samoa and New Caledonia — soon was 
scrapped. "The catastrophe of Midway 
definitely marked the turning of the tide 
in the Pacific War . . ." ^^ and from ar- 
rogant offense the Japanese soon turned 
to chagrined defense and ultimate defeat. 
U. S. plans for a first offensive already 
were well advanced, and the rest of 1942 
was destined to be a most gloomy period 
for the Japanese. 

In Battle That Doomed Japan^ Fuchida 
and Okumiya devote their final chapter to 
a scholarly and complete analysis of this 

Battle That Doomed Japan does not bear them 
out. The authors list only six planes lost in the 
Midway strike and 12 in combat air patrol. 

" Coral Sea and Midway, 75. 

"^ Ibid. See also Battle That Doomed Japan, 

"Battle That Doomed Japan, 231. 



defeat, and they end on an introspective 
note : 

In the final analysis, the root cause of Japan's 
defeat, not alone in the Battle of Midway but 
in the entire war, lies deep in the Japanese na- 
tional character. There is an irrationality and 
impulsiveness about our people which results in 
actions that are haphazard and often contradic- 
tory. A tradition of provincialism makes us 
narrow-minded and dogmatic, reluctant to dis- 
card prejudices and slow to adopt even necessary 
improvements if they require a new concept. 
Indecisive and vacillating, we succumb readily 
to deceit, which in turn makes us disdainful of 
others, ppportunistic but lacking a spirit of 
daring and independence, we are wont to place 
reliance on others and to truckle to superiors. 

Our want of rationality often leads us to confuse 
desire with reality, and thus to do things without 
careful planning. Only when our hasty action 
has ended in failure do we begin to think ration- 
ally about it, usually for the purpose of finding 
excuses for the failure. In short, as a nation, 
we lack maturity of mind and the necessary 
conditioning to enable us to know when and what 
to sacrifice for the sake of our main goal. 

Such are the weaknesses of the Japanese na- 
tional character. These weaknesses were re- 
fiected in the defeat we suffered in the Battle of 
Midway, which rendered fruitless all the valiant 
deeds and precious sacrifices of the men who 
fought there. In these weaknesses lies the cause 
of Japan's misfortunes.^" 

' Ibid., 247-248. 


The Turning Point: Guadalcanal 


Background and Preparations ^ 

Scarcely had Admiral Yamamoto pulled 
his Combined Fleet away from its defeat 
at Midway before the U. S. Joint Chiefs of 
Staff began reconsidering basic Pacific 
policy. They wanted an offensive which 
would aid containment of the Japanese ad- 
vances toward Australia and safeguard the 
U. S. communication lines to the Anzac 
area. As early as 18 February, Admiral 

' Unless otherwise noted the material used in 
Part VI is derived from 1st MarDiv FinalRept on 
the Guadalcanal Operation, Phases I through V, 
issued June-August 1943, hereinafter cited as 
FinalRept (with Phase No) ; action reports, war 
diaries, and journals of the various units which 
served with or as part of the 1st MarDiv ; Marine 
Air History ; Strategic Planning ; W. F. Craven 
and J. L. Gate (eds.). The Pacific: Guadalcanal 
to Saipan — The Army Air Forces in World War 
II (Chicago : University of Chicago Press, 1050) ; 
J. Miller, Jr., Ouadalcanal: The First Offensive — 
United States Army in World War II (Washing- 
ton: HistDiv, DA 1949), hereinafter cited as 
Miller, Ouadalcanal; S. E. Morison, The Struggle 
for Guadalcanal — History of United States Naval 
Operations in World War II (Boston: Little, 
Brown and Company, 1950), hereinafter cited as 
Struggle for Ouadalcanal ; Capt H. L. Merillat, 
The Island (Boston: Houghton Mifflin Company, 
1944), hereinafter cited as The Island; Maj J. L. 
Zimmerman, The Guadalcanal Operation (Wash- 
ington : HistDiv, HQMC, 1949) ; VAdm R. Tanaka 
with R. Pineau, "Japan's Losing Struggle for 
Guadalcanal," two parts, USNI Proceedings, July 
and August 1956, Copyright 1956 by the U. S. 
Naval Institute, hereinafter cited as Tanaka 
Article. Specific citations of material, in addition 
to direct quotations, taken from FinalRept have 
been noted where the information presented may 
be of special interest. 

Ernest J. King, Commander in Chief of 
the U. S. Fleet and Chief of Naval Opera- 
tions, told Chief of Staff George C. 
Marshall that he considered it necessary to 
garrison certain South and Southwest 
Pacific islands with Army troops ^ in prep- 
aration for launching U. S. Marines on an 
early offensive against the enemy .^ And 
shortly after the Battle of the Coral Sea, 
General MacArthur advanced plans for an 
attack against the Japanese at Rabaul. 
For this move he requested aircraft car- 
riers, additional troops, and more planes.* 
But Nimitz rejected this plan. His car- 
riers were too precious for commitment in 
waters so restricted as the Solomon Sea, he 
told the general. Besides, the admiral had 
a plan of his own. He wanted to capture 
Tulagi with one Marine raider battalion.^ 
Admiral King's reaction to this plan was 
initially favorable, but on 1 June he sided 
with Marshall and MacArthur that the 
job could not be done by one battalion. 
(See Map 11) 

^CominCh Itr to CofSA, 18Feb42 (located at 

' Specific mention of Marines for assault work 
came after Marshall questioned King's plans and 
asked why these FMF troops could not perform 
the garrison duty. CofSA Itr to CominCh, 
24Feb42; CominCh Itr to CofSA, 2Mar42 (lo- 
cated at NHD). 

* CinCSWPA msg to CofSA, 8May42 (located at 

= CinCPac Itr to CinCSWPA, 28May42 (located 
at NHD). 




But now time and the victory at Midway 
had improved the U. S. position in the 
Pacific, and on 25 June Admiral King ad- 
vised Nimitz and Vice Admiral Robert L. 
Ghormley, Commander of South Pacific 
Forces,* to prepare for an offensive against 
the Lower Solomons. Santa Cruz Island, 
Tulagi, and adjacent areas would be seized 
and occupied by Marines under CinCPac, 
and Army troops from Australia then 
would form the permanent occupation 
garrison/ D-Day would be about 1 

The task seemed almost impossible. 
Ghormley had just taken over his Pacific 
job after a hurried trip from London 
where he had been Special Naval Observer 
and Commander of U. S. Naval Forces in 
Europe; the 1st Marine Division, slated 
for the Solomon landing, was making an 
administrative move from the United 
States to New Zealand ; and Marshall and 
King continued to debate matters of com- 
mand. The general contested the Navy's 
right to command the operation. The area 
lay in the Southwest Pacific, Marshall 
pointed out, and so MacArthur ought to be 
in charge.^ 

Never mind arbitrary geography. King's 
reply seemed to say. The forces involved 
would not come from MacArthur, but from 
the South Pacific ; and King doubted that 
MacArthur could help the operation much 
even if he wanted to. The Southwest 

" Two days earlier a message from Nimitz gave 
Ghormley the Midway victory tally and suggested 
that the carriers now might be made available for 
support of . an operation against the Solomon 
Islands. ComSoPac War Diary, June 1942 
(located at NHD). 

' CominCh disp to CinCPac and ComSoWesPac- 
For, 25Jun42 (located at NHD). 

"CofSA Itr to CominCh, 26Jun42 (located at 

Pacific's nearest land-based bomber field 
was 975 miles from Tulagi. The command 
setup must be made with a view toward 
success, King said, but the primary con- 
sideration was that the operation be begun 
at once. He stated unequivocally that it 
must be under Nimitz, and that it could not 
be conducted in any other way.* 

The Joint Chiefs resolved this conflict 
on 2 July with issuance of the "Joint Di- 
rective for Offensive Operations in the 
Southwest Pacific Area Agreed on by the 
United States Chiefs of Staff." The direc- 
tive set the seizure of the New Britain-New 
Ireland-New Guinea area as the objective 
of these operations, but it broke this goal 
down into three phases designed to resolve 
the dispute between MacArthur and 
Nimitz. Phase One would be the seizure 
of the islands of Santa Cruz and Tulagi, 
along with positions on adjacent. islands. 
Nimitz would command this operation, 
with MacArthur concentrating on inter- 
diction of enemy air and naval activity to 
the west. And to remove MacArthur's 
geographic claim on the Phase One target 
area, the Joint Chiefs shifted the boundary 
between the general and Admiral Nimitz 
to place the Lower Solomons in the ad- 
miral's South Pacific area. MacArthur 
then would take command of Phase Two, 
seizure of other Solomon Islands plus posi- 
tions on New Guinea, and of Phase Three, 
the capture of Rabaul and adjacent bases 
in New Britain and New Ireland. Ques- 
tions of timing, establishment of task 
organizations, and arrangements for com- 
mand changes from*^ one area to another 
would be governed by the Joint Chiefs. 

Preparation of this directive in Wash- 
ington had prompted King's warning 

"CominCh Itr to CofSA, 26Jun42 (located at 



order which Ghormley received on 25 
June; and when the directive arrived in 
the South Pacific tlie force commander 
there already was making his plans for 
Phase One, which Washington labeled 
Operation WATCHTOWER. But, valid 
as was the Chiefs' of Staff determination 
to lose no time in launching this first offen- 
sive, problems facing Ghormley and Nim- 
itz were so grim that the pseudo code name 
for the undertaking soon became "Opera- 
tion Shoestring." 


Since Pearl Harbor the Japanese had 
expanded through East Asia, the Indies, 
and much of Melanesia to a gigantic line 
of departure which menaced Australia 
from the Indian Ocean to the Coral Sea. 
Lae, Salamaua, and Finschhafen on New 
Guinea's north coast had been occupied, 
and a force for the capture of Port 
Moresby — a New Guinea town just across 
the north tip of the Coral Sea from Aus- 
tralia's Cape York Peninsula — stood 
poised at Rabaul in the Bismarcks, a posi- 
tion taken by the Japanese on 23 January 

A month later the Japanese took Boug- 
ainville Island in the Northern Solomons, 
and on 4 May they took a 300-mile step 
down this island chain to capture Tulagi, 
which lay between the larger islands of 
Florida and Guadalcanal. This started 
the Japanese encirclement of the Coral 
Sea, a move that was thwarted by Admiral 
Fletcher in the naval battle that preceded 
the fight at Midway (see Part V, 
Chapter 1). 

Defeat of their fleet at Midway forced 
the Japanese to alter many of their ambi- 
tious plans, and on 11 July they gave up 
the idea of taking New Caledonia, Fiji, and 

Samoa. But if Admiral Yamamoto real- 
ized that the failure of his fleet at Midway 
spelled the doom of Japanese ambitions in 
the Pacific, U. S. fighting men were to meet 
a number of his countrymen who did not 
get the word, or who were bent on convinc- 
ing Yamamoto that he was wrong. Rabaul 
and Solomons positions grew stronger 
after the Battle of Midway, and reduction 
of Fortress Rabaul would occupy efforts of 
the Allied South Pacific forces for nearly 
two years. Operation WATCHTOWER, 
which turned out to be the landing against 
Tulagi and Guadalcanal, was the first 
Allied step toward Rabaul. 

In 1942 the Australian garrison at 
Tulagi consisted of a few riflemen of the 
Australian Imperial Force, some members 
of the Royal Australian Air Force, a mem- 
ber of the Australian Naval Intelligence, 
the Resident Commissioner, the civil staff, 
and a few planters and missionaries. Most 
of these people evacuated the area after a 
heavy Japanese air raid of 1 May, and the 
subsequent sighting by coastwatchers of 
enemy ships en route toward the Southern 
Solomons. Among those who remained in 
the Solomon area were the coastwatchers, 
courageous old island hands who now re- 
tired into the bush and hills from which 
they would observe Japanese movements 
and report regularly by radio to their 
intelligence center in Australia." 

The 3d Kure Special Landing Force 
made the Tulagi landing from the cruiser- 
mine layer Ohinoshima which flew the flag 
of Rear Admiral Kiyohide Shima. One 
group of these Japanese "marines" — a ma- 
chine gun company, two antitank gun 

" For an excellent account of the work of these 
men see Cdr E. A. Feldt, RAN, The Coastwatchers 
(New York and Melbourne: Oxford University 
Press, 1946), hereinafter cited as Coastwatchers. 



platoons, and some laborers — occupied 
Tulagi while a similar task organization 
from the Sd Kv/re Force went ashore on 
Gavutu, a smaller island nearby. They 
met no opposition, except that from Ad- 
miral Fletcher's planes in the action an- 
cillary to the Battle of the Coral Sea, and 
defensive installations were set up immedi- 
ately to protect the base construction and 
improvement work which soon got under- 
way. The Japanese set up coastwatcher 
stations on Savo Island at the northwest 
end of the channel between Florida and 
Guadalcanal, and on both tips as well as 
the south coast of Guadalcanal. 

Tulagi has an excellent harbor ,^^ and in- 
itial efforts of the Japanese landing force 
improved this and developed a seaplane 
base there. The enemy took no immediate 
steps to develop airfields, and a full month 
passed before surveying parties and 
patrols crossed the 20-mile channel to 
Guadalcanal where they staked out an 
air strip site on the plains of the Lunga 
River. They finished this survey late in 
June and began to grade a runway early in 

With a scrapping of the plans to occupy 
Samoa, Fiji, and New Caledonia, and with 
the importance of Rabaul thus increased, 
Japanese holdings in the Lower Solomons 
had gained in value. Tulagi with its ex- 
cellent harbor, and Guadalcanal with its 
broad plains suitable for airstrips, would 
be an important outwork to Rabaul. A 
new offensive likewise could be mounted 
from the Bismarck-Solomon positions, to 
erase the Coral Sea and Midway setbacks. 

Japan still had her eye on Port Moresby. 
The troops slated for that occupation 
already waited at Rabaul, and now a new 
fleet, the Eighth^ under Vice Admiral 
Gunichi Mikawa, was created to help look 
after this southern end of the Japanese 
conquest string. This fleet, with the help 
of aircraft from Rabaul and the Lower 
Solomons, would protect the ferrying of 
troops to Buna, and the subsequent over- 
land march of these troops across New 
Guinea's Owen Stanley Mountains to cap- 
ture Port Moresby. Thus Australia would 
be well blocked if not completely isolated ; 
and maybe if the Japanese did not think 
about the defeat at Midway the sting 
would just go away and everything would 
be all right again.^^ 


After reinforcing the Anzac lifeline (see 
Part II, Chapter 3), the U. S. began edg- 
ing toward its Solomon Islands target 
area. Near the end of March some 500 
Army troops from Major General Alex- 
ander M. Patch's America Division in 
New Caledonia went up to garrison Efate 
in the lower New Hebrides. On 29 March 
the 4th Marine Defense Battalion and Ma- 
rine Fighter Squadron 212, diverted from 
their deployment from Hawaii to Tonga- 
tabu, also landed on this island. These 
Marines and Army personnel built an air- 
strip for the fighter squadron. Naval 
forces also began to arrive during this 
time, and in April other elements of Ma- 
rine Air Group 23, parent organization 
of Squadron 212, came to the island. The 
prewar seaplane base at Vila, Ef ate's larg- 

""Earl Jellicoe, the British admiral who com- 
manded at Jutland, recommended after a South 
Pacific inspection trip following World War I 
that the Tulagi harbor be developed as a major 
fleet base. 

" For an idea of the pains taken by oflScial 
Japan to hide the facts of the Midway defeat, see 
the preface of Mitsuo Fuchida in Battle That 
Doomed Japan, xiii-xv. 



est town, was improved, and another such 
base was built in Havannah Harbor on 
the island's northwest coast. 

This was a hazardous and rather un- 
nerving extension of defensive lines for 
the meager American force of that period. 
The Japanese were just 700 miles to the 
north in the Lower Solomons, and this 
fact gave these New Hebrides islands and 
waters that same hostile, "creepy" feeling 
that members of the unsuccessful Wake re- 
lief expedition had sensed while on that 
venture deep into the enemy zone. The 
Japanese made a few air raids into this 
area, but the greatest opposition came from 
the anopheles mosquito. 

In the New Hebrides American troops 
of World War II had their first wholesale 
encounter with this carrier of malaria, and 
the field medical units were not prepared 
to cope with the disease in such propor- 
tions. Atabrine tablets were not yet avail- 
able, and even quinine was in short supply. 
By the end of April there were 133 cases 
of malaria among the 947 officers and men 
of the 4th Defense Battalion, and by the 
time of the Guadalcanal landing early in 
August the entire New Hebrides force re- 
ported the disease in even greater propor- 
tions. Medical units were dispatching 
requests for "an enormous amount of 
quinine." ^* 

In May both General Patch and Admiral 
Ghormley recommended that a force be 
sent even farther north, to EspiritjLSanto. 
An airfield there would put the Allied 
planes J,50LJuiles closer to the Solomons, 
and the force there would be a protective 
outwork for Ef ate until that first offensive 
started. Admiral Ghormley also recom- 
mended that the Ellice Islands, between 
the Samoan group and the Gilberts, be oc- 

' MedRept, 4th DefBn, March-August 1942. 

cupied as an additional outpost of the 
communication lines to Australia. This 
move was postponed, however, and it was 
not until October that Marines landed at 
Funafuti in the Ellice group. Espiritu 
Santo was immediately important to the 
Solomons operation, however, and on 28 
May a force of about 500 Army troops 
moved from Efate to the larger New He- 
brides island farther north. The first at- 
tempt of these troops to build an airfield 
there bogged down in a stretch of swamp 
and new outbreaks of malaria. 

By this time, plans for the WATCH- 
TOWER landings were firming up, and 
the effort at Espiritu Santo was reinforced 
so that the airfield would be completed in 
time. On 15 July a detachment from the 
4th Marine Defense Battalion went up to 
Santo with a heavy antiaircraft battery 
and an automatic weapons battery. The 
airfield was completed in time, but the 
Army troops and Marines were mostly 
walking cases of malaria by then. How- 
ever, important islands had been rein- 
forced, new garrisons formed to protect 
the communication lines, and these dis- 
placements toward enemy bases had been 
accomplished. The time had come to 
strike back at the Japanese. 


When Admiral Ghormley received the 
WATCHTOWER warning order on 25 
June, the 1st Marine Division, commanded 
by Major General Alexander Archer 
Vandegrif t, was en route from the United 
States to Wellington, New Zealand. The 
advance echelon had arrived on 14 June, 
and the rear was at sea. It would land on 
11 July. Until 26 June when information 
of the operation reached the division staff, 
Vandegrift had planned to continue 



training his division in New Zealand.^^ 
The division, the Marine Corps' major unit 
available for employment on such short 
notice, was understrength by about one 
third because of detachment of the rein- 
forced 7th Marines to S^moan duty. 

Army troops in the area, originally un- 
der Ghormley's command, could provide 
little more than moral support to the 
landings. This shoestring venture would 
not remove the need for garrison forces 
elsewhere in the South Pacific. Besides, 
Ghormley lost his direct control of these 
troops on 1 July when Major General 
Millard F. Harmon, USA, became Com- 
manding General, South Pacific Area, to 
head all Army forces in the theater. Even 
though Harmon would be under Ghorm- 
ley's command, the admiral at first dis- 
liked this command setup. But he later 
came to regard Harmon as one of the finest 
administrators and coordinators he had 
ever met.^® 

Admiral Ghormley's job in the South 
Pacific seemed almost to resemble that of 
a traffic director more than it did the role 
of a commander. According to plans, 
Nimitz would order task force command- 
ers, with their missions already assigned 
them, to report to Ghormley when they 
were going to carry out missions in his 
area. Ghormley then would direct these 
commanders to execute the missions Nimitz 
had assigned them, and he could not in- 
terfere in these missions except when tasks 

" Gen Vandegrift was under the impression 
that his division would not be called for combat 
duty prior to early 1943. 

" Adm R. L. Ghormley personal notes, n. d., 
hereinafter cited as Ghormley MS; Maj J. L. Zim- 
merman interview with Adm R. L. Ghormley, 
January 1949. 

were opposed by circumstances of which 
Nimitz was not aware.^' 

To complete the picture of command for 
WATCHTOWER, Rear Admiral Rich- 
mond K. Turner arrived from Washington 
on 18 July and reported to Ghormley as 
commander of the amphibious force. 
Ghormley, under Nimitz, was in over-all 
strategic command, but he would remain 
at his headquarters in Noumea. Admiral 
Fletcher would command the joint expe- 
ditionary force. But in practice Fletcher 
confined himself almost completely to pro- 
viding air cover from his carriers, and this 
left Turner, in addition to commanding 
Vandegrift and his division, in charge of 
almost everything else as well. 

This command setup which placed Van- 
degrift under the Navy's amphibious force 
commander rankled until nearly the time 
of the withdrawal of the 1st Marine Divi- 
sion from Guadalcanal fighting. It was 
not a case of small jealousy about control 
or any sort of petty peevishness on the part 
of either Vandegrift or Turner. Rather 
it was a clash of serious opposing convic- 
tions about how such an operation should 
be conducted. Turner and many other 
Naval authorities looked upon the landing 
force as just a detachment from the force 
afloat, and still connected to the Navy's 
amphibious force by firm command lines. 
That was a traditional view from an 
earlier age. 

But this was the beginning of a big new 
war, and Marines had experience and 
thinking time enough in amphibious 
matters to have definite studied opinions 
about how these intricate over-the-beach 
operations should be conducted when they 
reached the proportions which would be 

"ComSoPac War Diary, 9May42 (located at 



necessary in the Pacific. Vandegrift, 
faced with the task of putting these 
studied opinions and experiments into 
practice, wanted a clear-cut command 
right, free from any vestige of divided 
responsibility shared with the commander 
afloat. Once firmly established ashore, 
Marine opinion held, the landing force 
commander should command his own land 
operation. His training and position on 
the battlegi'ound made him more qualified 
for this job than was the amphibious force 
commander. It took some arguing, and 
this matter finally had to be taken to the 
top of military hierarchy, but the Navy 
eventually saw the point and agreed 
with it. 

Turner's second in command was Rear 
Admiral V. A. C. Crutchley, RN, whose 
covering force would include eight cruis- 
ers (three Australian and five U. S.) and 
fifteen destroyers (all U. S.) . These ships 
were to provide naval gunfire support and 
antiaircraft protection. In all, the naval 
contingent included three aircraft carriers 
with a strength of 250 planes; a number 
of light and heavy cruisers; two new 
battleships; and the available screening 
vessels and auxiliary craft. Transports 
and cargo vessels were at a premium, and 
would continue so for some time. 

In addition to the approximately 250 
carrier aircraft, Ghormley could muster 
only 166 Navy and Marine Corps planes 
(including two Marine squadrons — VMF- 
212 and VMO-251), 95 Army planes, and 
30 planes from the Royal New Zealand Air 
Force. These 291 aircraft — all unfortu- 
nately based beyond striking range of the 
target area — were under the command of 
Rear Admiral John S. McCain whose title 
was Commander Aircraft South Pacific. 

He was instrumental in bringing about the 
construction of the Espiritu Santo airfield 
and seeing that it was available for air- 
craft on 28 July, in spite of all the troubles 
which befell the force in the New Hebrides. 

VMO-251 came in to Noumea on 12 July 
on board the USS Heywood. The outfit 
barely had time to set up camp at Tontouta 
and uncrate its aircraft before it got the 
word to go up to that new field at Santo 
and back up the landing. On 2 August 
the unit began to arrive at this northern 
New Hebrides field, and within nine days 
Lieutenant Colonel John N. Hart had his 
squadron installed there with its sixteen 
F4F-3P long-range photographic planes. 
Hart still was short his wing tanks for 
long-range flying, however. These were 
finally flown out from Pearl Harbor and 
arrived on 20 August. 

MacArthur's contribution to the Guadal- 
canal operation consisted of about sixteen 
B-l7's which flew reconnaissance over 
the area west of the 158th meridian east 
(the boundary between the South and 
Southwest Pacific Areas for air search) 
and attempted to put a stopper on the 
enemy air from Rabaul. 

Thus Ghormley could rely only on the 
services of a small, highly trained striking 
force of fluctuating but never overwhelm- 
ing power. He had no assurances of re- 
serve ground troops, although plans were 
under way to release both the 7th and 8th 
Marines from their Samoan defense mis- 
sions,^^ and he had been advised that gar- 
rison forces would have to come from the 
troops already within his area on base de- 

" Relief for 7th Mar. was to leave the U. S. on 
20 July and that for the 8th Mar on 1 September. 
ComSoPac War Diary, June 1942 (located at 



fense duty." The 1st Base Depot had set 
up an advance echelon in Wellington on 
21 June, and other supply bases were to be 
established later at Noumea and Espiritu 

The general structure organized to em- 
ploy these resources against the Japanese 
was laid down in Nimitz' order to Ghorm- 
ley of 9 July, and Ghormley's Operation 
Plan 1^2 of 17 July 1942.^" 

Ghormley, exercising strategic com- 
mand, set up his organization in three 
main groups : 

Carrier Forces (Task Force 61.1), com- 
manded by Rear Admiral Leigh Noyes, 
was composed of elements of three task 
forces from Nimitz's area — 11, 16, and 18. 
It would include three carriers, Saratoga^ 
Enterprise, and Wasp, the fast new battle- 
ship North Carolina, five heavy cruisers, 
one so-called antiaircraft cruiser, and 16 

Amphibious Force (Task Force 61.2), 
commanded by Rear Admiral Richmond 
K. Turner, included the Marine landing 
force carried in 13 attack transports, four 
destroyer transports, and six cargo ships ; 
a fire support group of one antiaircraft 
and three heavy cruisers plus six de- 
stroyers; a screening group of one light 
and three heavy cruisers as well as nine 

" Adm King's effort to secure quick release for 
the assault troops was not successful. The 
Army's commitments to the European Theater 
were such that no units were available for such 
missions. Initially assured that air support and 
air replacements would be available, King was 
advised on 27 July by LtGen Joseph C. McNarney, 
acting Chief of Staff, that commitments in other 
areas would not permit further air reinforce- 
ments of the South Pacific — a dictum which 
King protested strongly. CominCh memo to 
CofSA, lAug42 (located at NHD). 

'° Ghormley MS, 54, 58. 

destroyers; and a small mine-sweeping 

Shore-Based Aircraft (Task Force 63), 
under Rear Admiral J. S. McCain 
(ComAirSoPac), included all aircraft in 
the area except those on carriers. 

Complicating this symmetrical structure 
was the tactical command role played by 
Vice Admiral Fletcher. He was in over- 
all command of TF 61 which included the 
forces of Noyes and Turner. 

Ghormley called for a rehearsal in the 
Fiji area and directed that all task force 
commanders arrange to hold a conference 
near the rehearsal area. He himself would 
move from Auckland to Noumea about 1 
August in order to comply with his orders 
to exercise strategic command within the 
operating area.^^ 

By this time the plans and orders were 
formed, the target selected, the forces or- 
ganized, and the Navy given leeway to op- 
erate without poaching in the territory of 
the Southwest Pacific. Only the detail of a 
landing date remained unsettled. Vande- 
grift pointed out to Ghormley that the 
late arrival of his second echelon, and a 
stretch of bad weather, had so complicated 
his loading problem as to make it impos- 
sible to meet the date of 1 August. Ghorm- 
ley and Nimitz agreed that an additional 
week was needed, and King consented to 
postpone the landing until 7 August. 
King warned, however, that this was the 
latest date permissible and that every ef- 
fort should be made to advance it. 

[accumulation of 

From an intelligence point of view, the 
Guadalcanal-Tulagi landings can hardly 

Ibid., 59. 



be described as more than a stab in the 
dark. When General Vandegrift received 
his initial warning order on 26 June 1942, 
neither his staff nor the local New Zealand 
authorities had more than the most gen- 
eral and sketchy knowledge of the objec- 
tive area or the enemy's strength and dis- 
position, and there was but a month avail- 
able before the scheduled date of mount- 
ing out, 22 July. 

As in the case with most tropical back- 
waters, the charting and hydrographic in- 
formation was scanty and out of date. 
Lieutenant Colonel Frank B. Goettge, In- 
telligence Officer of the 1st Marine Divi- 
sion, therefore set out to locate traders, 
planters, shipmasters, and a few miners 
who had visited or lived at Guadalcanal 
or Tulagi. A number of likely sources re- 
sided in Australia, and while his subordi- 
nates tabulated the formal data available, 
Goettge left for Australia on 2 July. He 
returned to New Zealand on the 13th. 

Ijong after the conclusion of the cam- 
paign, it was learned that Colonel 
Goettge's efforts deserved better success 
than they had enjoyed. During his hur- 
I'ied trip to Australia, he arranged with the 
Southwest Pacific Area for maps to be 
made from a strip of aerial photographs 
and to be delivered prior to the sortie of 
the 1st Marine Division. Themaps were 

made^but were not received because j of, 
certain oversfghts and confusiOTi m mount- 
ing out the division. 

From Buka and Bougainville in the 
north, the Solomons form a double column 
of islands streaming southeast between 
latitudes five and twelve degrees south. 
Looking northwest toward Bougainville, 
the large islands on the right are Choiseul, 
Santa Isabel, and Malaita. In the column 
to the left are the islands of the New Geor- 

gia group, the Russells, Guadalcanal, and 
San Cristobal. Buka and Bougainville at 
outbreak of the war were part of the Aus- 
tralian Mandated Territory of New 
Guinea ; the remainder of the double chain 
formed the British Solomon Islands Pro- 
tectorate. In all, the islands number sev- 
eral hundred, with some 18,600 square 
miles of land area. (See Map 11) 

Florida, the largest island of the Nggela 
Group, lies between Malaita and Guadal- 
canal; and between the northern tips of 
Guadalcanal and Florida is the small, 
nearly-conical island of Savo. Indispens- 
able Strait separates Florida from neigh- 
boring Malaita to the east, and the twenty- 
mile-wide strait between Florida and 
Guadalcanal to the south is known gen- 
erally as Sealark Channel. (See Map 13, 
Map Section) 

Nestled into the northwest rim of a 
jagged bight in Florida's south coast lies 
Tulagi, seat of the British Resident Com- 
nxjssioner. Tulagi Harbor, the water be- 
tween the two islands, is the best anchor- 
age in the Southern Solomons. 

In the middle of Florida's bight, gen- 
erally east southeast of Tulagi, lie the 
smaller causeway-connected islands of 
Gavutu and Tanambogo. Gavutu was the 
local headquarters of Lever Brothers 
which operated coconut plantations in the 
_ area, and this island, as well as Tanambogo 
and Tulagi, possessed some docks, jetties, 
and other developments for shipping, 
management, and copra processing. 

Mostly volcanic in origin and lying 
within the world's wettest area, the Solo- 
mons are jagged, jungle-covered, and 
steamy with humid tropical heat. Lofty 
peaks and ridges cross-faulted by volcanic 
action and dramatic erosion cuts from 
swift rivers chop the islands into conflict- 





ing terrain that became a nightmare for 
military operations. 

Guadalcanal, some 90 miles in length 
and about 25 miles wide, presents a varied 
topography ranging from plains and foot- 
hills along the north coast to a mountain 
backbone dropping rapidly to the south 
coast. Rainfall is extremely heavy, and 
changes in season are marked only by 
changes in intensity of precipitation. 
This, together with an average tempera- 
ture in the high 80's, results in an un- 
healthy climate. Malaria, dengue, and 
other fevers, as well as fungus infections, 
afflict the population. 

Rivers are numerous and from the mili- 
tary point of view may be divided arbi- 
trarily into two classes. The first of these 
is the long, swift, relatively shallow river 
that may be forded at numerous points. 
Generally deep only for a short distance 
from its mouth, it presents few problems 
in the matter of crossing. Examples of 
this type on Guadalcanal are the Tenaru, 
the Lunga, and the Balesuna. The second 
type is that of the slow and deep lagoon. 
Such streams are sometimes short, as in 
the case of the Ilu River, and some lagoons 
are merely the delta streams of rivers of 
considerable size, as in the case of the 
Matanikau. This type, because of depth 
and marshy banks, became a military 

Although such accumulation of data af- 
forded much enlightenment beyond the 
little previously known, it included corre- 
sponding minor misinformation and many 
aggravating gaps, for detailed informa- 
tion in a form suitable for military oper- 
ations was mainly lacking. 

In spite of the number of years which 
had elapsed since initiation of the system- 
atic economic development of the islands 

by the British, not a single accurate or 
complete map of Guadalcanal or Tulagi 
existed in the summer of 1942. The hy- 
drographic charts, containing just suffi- 
cient data to enable trading schooners to 
keep from grounding, were little better, 
although these did locate a few outstand- 
ing terrain features of some use for mak- 
ing a landfall or conducting triangulation. 
Such locations were not always accurate. 
Mount Austen, .for example, was assigned 
as an early landing objective, but the land- 
ing force discovered that instead of being 
but a few hundred yards away from the 
beaches, the motmtain actually lay several 
miles across almost impassable jungle.^ 

Aerial photographs would have been a 
profitable source of up-to-date informa- 
tion, but the shortage of long-range air- 
craft and suitably located bases, and the 
short period available for planning, com» 
bined to restrict availability of aerial 
photos in the quantity and quality nor- 
mally considered necessary. 
[ Perhaps the most useful photographic 
sortie carried out prior to the Guadal- 
canal-Tulagi landings was that under- 
taken on 17 July by an Army B-17 air- 
craft in which Lieutenant Colonel Merrill 
B. Twining, assistant operations officer of 

22 This was the so-called "Grassy Knoll" as- 
signed to the 1st Marines. Guadalcanal residents 
described it as lying virtually within the perim- 
eter area ultimately occupied and defended by 
Gen Vandegrift, whereas its true location was 
six miles to the southwest. This discrepancy, 
unexplained for years, has given much cause for 
speculation to historians of the Guadalcanal cam- 
paign, some of whom have raised the question 
whether Mount Austen was really the "Grassy 
Knoll" which Goettege's informants had in mind. 
This school of thought suggests that a feature 
within the described limits which might answer 
that description would have been the high ground 
later to become known as Edson's Ridge. 




i=s»?kr$^fA:!2^j?!:L"3^ ,,.^jm^^^o4o ^.....i s..i...^o^^3 c«^^.^.^«^ 


lU Jul»,l9'»2 

Special Skatch ■uknlttad, uid 
■ollad br lUp; lorth Cout of 
ileaoal blaad, B.r. XtSJ.OitO ; 

S^atut» Miles 


SFKUl BOTZ: Thl* asp mi raprodaead froa 8p«clal Sketch M^ dram froa tha Inforsatlon auppllad 
by a B&Q thorouctily faalllar with tha tarraln ahovn. It la aa approzlaata pictorial rapraacat- 
atlOD drsvn froa tbla paraon's aaaory. It la not to ba ooaatruad aa balac aa acpvata aap. 

: 8 T M B L S : 


'"' (»ASS PUIHS (TBICI: >» 10 6 FT. EIOB) . 

jfe 4 SWAMP. ^Cii, C^^:? noiLB. 



-•— mcE. (IS tulr old) . 

miMSs. 3 va unit, on iut TBikiric . 

OKtM. UnnU. TBUIT aod ILD, ara FlaotaUo'..r. 

CRUDE SKETCH MAP used in the planning and early operational phases of the Guadalcanal campaign 
by units of the 5th Combat Team; it is an adaptation of a map prepared by the D-2 Section and typifies the 
scarcity of reliable terrain information available to the 1st Marine Division when it left New Zealand. 



the 1st Marine Division, and Major Wil- 
liam B. McKean, member of the staff of 
Transport Squadron 26, conducted a per- 
sonal reconnaissance of the landing areas. 
They assured General Vandegrift that 
the Lunga beaches appeared suitable for 

The coastal map of Guadalcanal finally 
adopted by the 1st Marine Division (and 
employed, with such corrections as could 
later be developed, through the entire 
campaign) was traced from an aerial strip- 
map obtained by Colonel Goettge on his 
mission to Australia. It was reasonably 
accurate in general outline, but contained 
no usable indications of ground forms or 
r elevations. The Goettge map was supple- 


mented by iaerial photos of Tulagi, Gavutu, 
and Tanambogo Islands, and these con- 
stituted the sum of the Marines' knowledge 
of Tulagi and Guadalcanal prior to the 

Information concerning the enemy's 
strength, dispositions, and activities was 
collected by the U. S. planners from coast- 
watcher reports.^^ Strength figures were 
by no~nieahI~a^ definite or convincing as 
were the factual accounts of the defenses. 
Various intelligence estimates, prepared 
during July, gave figures as high as 8,400. 
Admiral Turner's Operation Plan A3-42, 

"HistSec, HQMC interview with Col W. B. 
McKean, 18Feb48. 

" Two aerial photos, taken on 2 August by a 
ComAirSoPac B-17 and developed aboard the 
USS Enterprise, were forwarded to Division 
Headquarters. They showed Tulagi defensive 
positions in sharp detail, and verified the reports 
of coastwatchers about the rapidly approaching 
completion of the airstrip in the Lunga plains. 

" "The invaluable service of the Solomon 
Islands coastwatching system . . . cannot l^e 
too highly commended." FinalRept, Phase I, 
Annex E, 2. 

issued at the Koro Island rehearsals in the 
Fijis on 30 July, estimated that 1,850 
enemy would be found on Tulagi and 
Gavutu-Tanambogo, and 5,275 on Guadal- 
canal. Both figures were high. A count 
of enemy dead in the Tulagi and Gavutu 
area placed the number of defenders at 
about 1,500 (including 600 laborers), 
while study of positions, interrogation of 
prisoners, and translation of enemy docu- 
ments on Guadalcanal proper indicated 
that about 2,230 troops and laborers had 
been in the Lunga area at the time of the 
Marines' landing. 

Close and determined combat was antici- 
pated with these forces; and on 17 July, 
Admiral Nimitz notified Admiral King 
that it would be unsafe to assume that the 
enemy would not attempt to retake the 
area to be attacked, and that, if insufficient 
forces were assigned, the Marines might 
not be able to hold on.^" 


For the dual landing operation. General 
Vandegrift divided his organization into 
two forces. The units landing on the 
Florida side of Sealark Channel (Group 
Yoke) were to be commanded by Brigadier 
General William H. Rupertus, the as- 
sistant division commander (ADC), while 
Vandegrift himself would exercise com- 
mand over Group X-Ray landing at 
Lunga Point. ( See Map 13, Map Section ) 

It was expected that the Florida-side 
landings would be more severely contested 
by the Japanese, and to that landing group 
the general assigned his best-trained units : 
the 1st Marine Raider Battalion, com- 

'"CinCPac disp to CominCh, 17Jul42 (located 
at NHD). 



manded by Lieutenant Colonel Merritt A. 
Edson ; the 1st Parachute Battalion of 
Major Robert H. Williams; and the 2d 
Battalion, 5th Marines under Lieutenant 
Colonel Harold E. Rosecrans, all with 
their reinforcing units attached. Edson 
would be the commander of the Tulagi 
landing force; Williams the commander 
at Gavutu-Tanambogo. The Guadalcanal 
group included Colonel Clifton B. Cates' 
1st Marines and Colonel Leroy P. Hunt's 
5th Marines (less 2/5), both reinforced, 
plus the balance of the division special and 
service troops.^^ 

The Tulagi plan called for the 1st 
Raider Battalion and 2d Battalion, 5th 
Marines to land in column on the island's 
south coast, turn east, and attack down 
the long axis of the island. This would 
be followed by 1st Parachute Battalion 
landings on Gavutu and Tanambogo, and a 
two-company sweep along Florida Island's 
coast line fronting Tulagi Bay. (See Map 
15, Map Section) 

The Guadalcanal scheme envisaged 
landing the 5th Marines (less 2d Bn) 
across a beach some few hundred yards 
east of the Lunga Point area where the 
Japanese were expected to be concentrated, 
and there to establish a beachhead. The 
1st Marines then would come ashore in a 
column of battalions and pass through this 
perimeter to take Mount Austen. The 

" At this time, the later-used phrase "regi- 
mental combat team" (RCT) had not come into 
uniform use. This we would currently style a 
"regimental landing tedm" (RLT). What 
Guadalcanal Marines labeled a "combat group" 
included a rifle regiment with its direct-support 
artillery battalion, engineers, signal, medical, and 
other supporting elements. Within the so-called 
combat groups, similar battalion-sized aggrega- 
tions were designated combat teams. This usage 
will be followed throughout this narrative. 

primary goal was to establish a beachhead 
in an area not strongly defended.^* 

To make up for the division's manpower 
shortage caused by the detached duty of 
the 7th Marines in Samoa, Admiral King 
on 27 June had proposed that Vandegrift 
be allotted the 2d Marines of the 2d Ma- 
rine Division. Accordingly this unit (re- 
inforced) sailed combat loaded from San 
Diego on 1 July.^^ The regiment would be 
the landing force reserve.^" 

While staff planners contemplated a tar- 
get area nearly as unfamiliar to them as 
the back side of the moon, other members 
of the landing force wrestled the monu- 
mental chore of preparing for the move- 
ment to combat. "Seldom," General Van- 
degrift said later, "has an operation been 
begun under more disadvantageous cir- 
cumstances." ^^ 

When the decision to land on enemy 
beaches reached the 1st Marine Division, 
the command post and the 5th Marine 
Regiment were in Wellington, New Zea- 
land; the 1st Marines and the 11th Ma- 
rines, less two of its battalions, were at sea 
en route to New Zealand ; service and spe- 
cial troops were split between the forward 
and rear echelons ; the 2d Regiment was on 

'*lst MarDiv OpOrd 7-42, 20Jul42. See Ap- 
pendix E. 

^ Original plans called for this unit to carry 
out projected landings at Ndeni in the Santa 
Cruz Islands. Needless to say, these were never 
made, although occupation plans for that island, 
always involving Marine forces, continue to ap- 
pear in Adm Turner's record until October 

'" "It is most desirable that 2d Marines be re- 
inforced and combat unit loaded and ready upon 
arrival this area for employment in landing op- 
erations as a reinforced regimental combat 
team." ComSoPac War Diary, 27Jun42 (located 
at NHD). 

" FinalRept, Phase V, 1. 



the way from San Diego to the South Pa- 
cific ; the 1st Raider Battalion was in Sa- 
moa, and the 3d Defense Battalion was in 
Hawaii. Preliminary plans and moves 
had to assemble these widely scattered 
units into a fighting force which could 
make an amphibious landing, one of the 
most intricate of military maneuvers. 
From the understanding that it would be 
the nucleus for the buildup of a force 
which would be trained for operations 
which might come late in 1942, the 1st 
Marine Division had to shift at once into 
hurried preparations to mount out for 

Most of the ships transporting units of 
the division had been loaded organization- 
ally for the voyage to New Zealand, but 
for the proposed amphibious assault the 
supplies had to be reshuffled and ships 
combat loaded so that items first needed in 
the fighting would be readily at hand in 
the holds. The reloading and re-embark- 
ing of Combat Group A (5th Marines, 
reinforced) went smoothly, uncomplicated 
by the necessity for simultaneous unload- 
ing and reloading which plagued the rear 
echelon. The group began embarkation on 
2 July and remained on board its transport 
to await the arrival of the rear echelon. 

This second echelon arrived on 11 July, 
and had eleven days to empty and reload 
its ships. No troops were disembarked ex- 
cept those who were to remain in New 
Zealand as rear echelon personnel. All 
others, who already had been in cramped 
quarters during the long trip across the 
Pacific, were put to work in eight-liour 
shifts, and parties of 300 men were as- 
signed to each ship.^^ 

Aotea Quay at Wellington was the scene 
of this squaring away. It was inadequate 
in all ways save that it could accommodate 
five ships at a time. Labor difficulties with 
the highly unionized stevedores resulted 
in the entire task being undertaken and 
carried through by Marines. Because of 
security regulations, no appeal to patriot- 
ism could be made to the regular dock 
workers since care was taken to have civil- 
ians believe that all the flurry was merely 
preparation for a training exercise. Dock- 
side equipment was meager, and there was 
no shelter close at hand. 

As the gear began to be juggled from 
ship to dock and back again, a cold, wet 
"southerly" settled down to lash New Zea- 
land. But in spite of the weather, work 
had to continue around the clock. Carton- 
packed food and other supplies "deterior- 
ated rapidly," the division later reported 
by way of an understatement, and the 
morale of troops followed the direction of 
the down-slanting rain. 

On the dock, cereal, sugar, and other ra- 
tions mushed together with globs of brown 
pulp that once had been cardboard boxes. 
A great number of wet cartons that were 
rushed to the hopeful safety of wool ware- 
houses later gave way under the weight of 
stacking. Lieutenant Colonel Randolph 
McC. Pate and his logistics section had a 
herculean task in managing this unloading 

'^ The passage from the United States to New 
Zealand had been particularly trying for the 
officers and men on board the Ericsson, a com- 

mercial ship under charter. Lack of proper 
food, and use of oil substitutes for shortening, re- 
sulted in loss of weight of as high as 23 pounds 
per man. Two meals only were served during 
the greater part of the passage, and one of 
these often consisted of soup or soup and bread. 
Medical officers estimated the daily calorie con- 
tent of meals at less than 1,500. The ship's per- 
sonnel enjoyed a full and well balanced diet 
during the same period. LMd., Phase I, Annex 
M, 1. 



and reloading. Transport quartermasters 
of the various ships supervised work on 
board while a relay of officers from the 
division took charge of the eight-hour 
shifts dockside. The New Zealand Army 
furnished 30 flatbed lorries and 18 ten- 
wheelers to transfer fuel, small-arms am- 
munition and explosives to dumps several 
miles away. 

There was not enough hold space for all 
the division motor transport. Most of the 
quarter- and one-ton trucks were put on 
board, but 75 percent of the heavier rolling 
stock was set aside to stay with the rear 
echelon that would be left behind when the 
division sailed for the Solomons. 

Engineers loaded what little dirt-mov- 
ing equipment they owned, but it was so 
meager that they hoped the Japanese 
would have most of the airfield built by 
the time it was captured. The engineer 
battalion also loaded bridging material, 
demolitions, and all available water sup- 
ply equipment. No major construction 
was contemplated in early phases of the 
operation, however, and equipment and 
siipplies for such work were not taken. 

With the D^ or an assistant in constant 
touch with the dockmaster, order began to 
appear from the chaos of strewn gear, 
swelling cereal, wilting cardboard cartons, 
and frayed tempers. But there were still 
serious problems. Supplies and equipment 
piled on docks often made it difficult for 
trucks to negotiate the narrow passages to 
reach all areas of the stacked gear, and 
during this mounting out certain modifica- 
tions of the logistical plan became neces- 
sary. As finally loaded, the Marines car- 
ried 60 days supplies, 10 units of fire for 
all weapons, the minimum individual bag- 
gage actually required to live and fight, 
and less than half the organic motor trans- 
port authorized. 

Regardless of the difficulties, however, 
the force sailed as scheduled at 0900 on 
22 July, under escort of cruisers of Ad- 
miral Turner's Task Force 62.*^ General 
Vandegrift, despite his request for a vessel 
better suited in communications and ac- 
commodations, had been directed to em- 
bark his command post in the USS 


In accordance with orders received from 
Nimitz on 1 July,^* Ghormley had directed 
that all forces involved in the assault make 
rendezvous at a position south of Fiji, out 
of sight of land so that there would be no 
chance of observation by enemy agents and 
no chance that an inadvertent tip-off would 
be made by friendly observers.^^ At that 
point there would be a conference between 
the commanding officers who had not as 
yet been able to discuss in person the vari- 
ous aspects of the operation. 

The components of the assault force not 
previously in New Zealand with the divi- 
sion were converging upon the rendezvous 
point from many directions. Colonel John 
M. Arthur's 2d Marines (reinforced), em- 
barked in the Crescent City. President 
Adams, President Hayes, President Jack- 
son, and Alhena, steamed south under es- 
cort of the carrier Wasp and a destroyer 

^ Adm Turner assumed the title of Commander, 
Amphibious Force South Pacific on his arrival in 
the South Pacific on 24 July. 

'' CinCPac OpOrd 34-42, cited in CinCPac War 
Diary, July 1942. By terms of this order also, 
Fletcher. CTF 11, had been ordered to assume 
command of the combined task forces at the 
rendezvous, and Ghormley had been put in com- 
mand of the operation. 

^Ghormley M 8, 04. 



EQUIPMENT FOR THE 1ST MARINE DIVISION, including tanks and amphibian tractors, is 
unloaded in New Zealand preparatory to the Guadalcanal operation. (USN 11526) 

^.iP ■«;^,^^^ 

MARINE RAIDERS and the crew of the submarine Argonaut line the deck of the vessel as it returns to 
Pearl Harbor after the Makin Island raid. ( USM 13859) 



screen.''® The 1st Raider Battalion, in the 
four destroyer transports of Transport 
Division 12, had been picked up at 

The 3d Defense Battalion (Colonel 
Robert H. Pepper) on board the USS 
Betelgeuse and Zeilin was en route from 
Pearl Harbor where it had been stationed 
since the outbreak of the war. It would 
meet the remainder of the force on 2 
August. The Carrier Force, built around 
the Saratoga and the Enter-prise^ with 
Fletcher flying his flag in the former, like- 
wise was on its way from Pearl Harbor. 
Rendezvous was made as planned, at 1400 
on 26 July, some 400 miles south of Fiji. 
The conference convened at once on board 
the Saratoga. Ghormley, unable to attend, 
was represented by his Chief of Staff, Rear 
Admiral Daniel J. Callaghan and his 
Communications Officer, Lieutenant Com- 
mander L. M. LeHardy. 

The conference pointed up several seri- 
ous problems. General Vandegrif t learned 
he would not have adequate air and surface 
support for the completion of the unload- 
ing phase of the operation. Fletcher 
wanted to retire within two days after the 
landing, and this meant that transport 
shipping would have to clear out within 
an unreasonably short period. The Marine 
general also learned that the 2d Marines, 
counted as his reserve, actually would be 
used for the proposed operation at Ndeni 
in the Santa Cruz islands. Admiral Cal- 
laghan reported Fletcher's retirement 
plans to Ghormley: "This sounds too 
sanguine to me," Callaghan reported, "but 
they believe it can be done. . . . AKs 
[cargo ships] may not be unloaded for 

three or four days." ^^ Ghormley, too, be- 
lieved that the ships could not be pulled 
out that soon. 

Landing rehearsals at the island of Koro 
in the Fijis were conducted from 28 
through 30 July, but Vandegrift labeled 
them a waste of time and effort. "A com- 
plete bust," he observed later .^* Necessity 
for conserving landing craft made it im- 
possible to conduct the practice landings 
in a realistic way, although the men in- 
volved were given additional training in 
debarkation,^® and attack force ships were 
able to practice their gunfire support. 

On 31 July, as night was falling, the 
ships weighed anchor and departed from 
Koro. The carrier task force proceeded 
north and west while the transports and 
their screen plodded steadily toward the 
Solomons. Almost 19,000 Marines. were 
embarked in the 19 transports and four 

All circumstances favored the advanc- 
ing convoy. Weather conditions during 
the final two days were extremely favor- 
able: sky topped by a low ceiling and 
winds gusty with intermittent rain squalls. 
There was no sign of enemy aircraft or 
submarines, and no indication that the 
approach was observed. In fact, enemy 

°° The regiment had been on board since 1 June, 
lying In the harbor of San Diego. 

'' Ghormley MS, 67. 

** Statement at Princeton, N. J., 12Mar48. 

"Gen Vandegrift noted at the time that the 
precious landing craft were not in the best of 
condition in any event — 12 of them were inoper- 
ative on one ship alone. Gen A. A. Vandegrift 
Itr to CMC, 4Feb49. 

*° A seemingly irreconcilable discrepancy of 
figures between those of the Amphibious Force, 
South Pacific, and the 1st Marine Division pre- 
vents a wholly accurate statement as to the num- 
ber of troops embarked then or landed subse- 
quently. The amphibious force lists a figure of 
18,722, while the division records list, variously, 
19,546 or 19,105. 



patrol planes were grounded at Rabaul on 
5 and 6 August because of bad weather." 

The convoy headed generally west from 
Fiji and well to the south of the Solomons 
chain. The course gradually shifted to 
the northward, and the night of 6-7 
August found the entire group of ships 
due west of the western extremity of 

Task Force 62, commanded by Admiral 
Turner, was divided into two Transport 
Groups. Transport Group X-Ray (62.1) 
commanded by Captain Lawrence F. Reif- 
snider, with the Guadalcanal forces em- 
barked, consisted of four subgroups, as 
follows : 

Transdiv A : Fuller, American Legion, Bellatrix. 
Transdiv B : McCaivley, Barnett, Elliot, Libra. 
Transdiv C : Hunter Liggett, Alchiia, Fomal- 

haut, Betelgeuse. 
Transdiv D : Crescent City, President Hayes, 

President Adams, Alhena. 

Transport Group Yoke (62.2) commanded 
by Captain George B. Ashe, and carrying 
the assault troops for the Tulagi landing, 
consisted of the following subgroups : 

Transdiv E : Neville, Zeilin, Heywood, President 

Transdiv 12 : Calhoun, Oregory, Little, McKean. 

(the destroyer transport group ).'^ 

At 0310, 7 August, the force was directly 
west of Cape Esperance with an interval 
of six miles between groups and a speed 
of 12 knots. Transport Group X-Ray 
steamed in two parallel columns of eight 

and seven ships each with a distance of 
750 yards between ships and an interval 
of 1,000 yards between columns. The 
rugged outline of the Guadalcanal hills 
was just visible to starboard when the 
course was shifted to 040°, and a few- 
minutes later the two groups separated 
for the completion of their missions. 
X-Ray, shifting still further to starboard, 
settled on course 075°, which took it along 
the Guadalcanal coast, while Yoke, on 
course 058°, crossed outside Savo Island, 
toward Florida. The final approach to 
the transport area was made without inci- 
dent, and there was no sound until, at 0614, 
the supporting ships opened fire on the 

There are some indications that the 
Guadalcanal operation on D-Day morning 
was something of a minor Pearl Harbor 
in reverse for the Japanese. A recent 
study of Japanese wartime messages indi- 
cates the enemy was aware that a U. S. 
force had sortied from Hawaii. Warn- 
ings were issued to Central Pacific out- 
posts ; Rabaul and points south were to be 
notified for information only. Com- 
mander of the Japanese Twenty -Fourth 
Air Flotilla (Marshall-Gilbert-Wake 
area) relayed his warning message south 
the next morning — at 0430 on 7 August. 
It was too late. Less than an hour later 
he received Tulagi's report that the U. S. 
striking force had been sighted in Sealark 
Channel at 0425.** 

"Cdr J. Shaw Itr to Maj J. L. Zimmerman, 
February 1949. 

" CTF 62 OPlan A3-42, 30Jul42. 

*^ CTG 62.1 ActRept 7-9Aug42, 3. 
"Capt. E. T. Layton, USN, Itr to HistBr, 
G-3, HQMC, June 1955. 


Guadalcanal, 7-9 August 1942 


When Task Groups X-Ray and Yoke 
separated northwest of Cape Esperance at 
0240, the former group made for the Red 
Beach transport area off Guadalcanal in a 
double column at 12 knots. No enemy ac- 
tivity was observed, and the preliminary 
naval bombardment of the coastal area, 
which began at 0613, aroused no response. 
The X-Ray shipping reached its transport 
area at 0645 and began to lower the land- 
ing craft. Across the channel, Group 
Yoke likewise arrived at its assigned area 
off Tulagi without incident at 0630 and 
straightaway got the word from Captain 
Ashe that H-hour would be 0800. The 
units slated for Florida Island would hit 
their beaches first, as will be described in 
the next chapter. 

The division's command post in the Mc- 
Caxdey broke radio silence at 0519, and 
eight minutes later General Vandegrift 
set the H-hour for his side of the landing 
at 0910. The bombardment ships worked 
through their fire plans, and then as news 
of the successful landings on Florida and 
Tulagi reached Vandegrift, the first waves 
of assault troops moved toward the beach. 
(See Map 14, Map Section) 

Three planes from the Astoria flew liai- 
son missions in the Guadalcanal area while 
three from the Vincennes performed the 
same duty above Tulagi. An additional 
three aircraft, from the Quincy, were 
available for artillery spotting over Guad- 
alcanal. During the ship to shore phase, 

these aircraft marked the beach flanks 
with smoke to assist naval gunfire and to 
guide the landing boats. Vandegrift and 
his division air officer held this use to be 
unwarranted and unnecessary. 

But Admiral Turner considered it neces- 
sary to "accurately mark the extremities 
of the landing beaches" as directed by the 
operation order, and he marked them for 
twenty minutes. The planes made eight 
runs at extremely low altitudes, four runs 
on each beach extremity. Vandegrift 
pointed out that this would result in a 
serious if not complete loss of planes if 
the beaches were defended — this loss at a 
time when aircraft are critically needed 
as "eyes" to gain information about the 
progress of a landing. 

Actually the liaison planes over Guadal- 
canal's random clouds and splotchy jungle 
furnished Vandegrift precious little infor- 
mation. It was not the fault of the pilots, 
however, since there was very little to 
see anyway. In the tense period of this 
first landing on a hostile beach, the sins 
were more often those of commission 
rather than omission. One pilot reported 
"many enemy troops" only to admit, under 
questioning for more explicit information, 
that his "troops" were, in fact, cows. 

Other than the cows there still were no 
signs of activity around Lunga at 0859, 
11 minutes before H-hour, when an ob- 
servation plane from the Astoria reported 
that no Japanese could be seen in that 
area. But 15 minutes later the same pilot 
spotted some trucks moving on the Lunga 






,. '^' 

^^^■K^SPiSP^^^' ; 

.. ',^^^';^ 


wtmwtifflp^ffigiifTr* Ijwijjjiii^ 

i. - ^^'i^Bia" 

'"j^^-4 ..-^HgiL.^ -^^laiiBilaitehfc. -. ^8i9^^^H9H 


.j3K'.- 'i^., -'^ '*i- 'hK "•"'.'•i-.^ (T^BP^T^TTjfflEE^i'- **'■ 


'^wT' -ii^amnwtpwa*'?^ 

'-■■'.,. %."^ 





THE ORIGINAL HENDERSON FIELD, target of the 1st Marine Division's assault at Guadalcanal, 
as it appeared shortly after its capture. ( VSMC 108547 ) 


TRANSPORTS AND CARGO SHIPS STAND OFFSHORE as vitally needed supplies for the 1st 
Marine Division are manhandled by members of the Shore Party. (USMC 51369) 



airfield several thousand yards west of 
the landing beach. 

Meanwhile the 5th Marines (less 2d Bat- 
talion) had crossed its line of departure 
and moved into the 5,000-yard approach 
to the beach. Naval gunfire lifted inland 
as the craft neared the shore, and minutes 
later, at 0910, the assault wave hit the 
beach on a 1,600 yard front and pushed 
into the sparse jungle growth beyond. 
With Lieutenant Colonel William E. Max- 
well's 1st Battalion on the right (west) 
and Lieutenant Colonel Frederick C. Bie- 
bush's 3d Battalion on the left, the beach- 
head expanded rapidly against no opposi- 
tion. A perimeter some 600 yards inland 
soon established a hasty defense. The line 
anchored on the west at the Tenaru River, 
on the east at the Tenavatu River, and 
reached on the south an east-west branch 
of the Tenaru.^ 

Regimental headquarters came ashore 
at 0938 to be followed two minutes later 
by heavy weapons troops. Landing of the 
reserve regiment. Colonel Cates' reinforced 
1st Marines, already was underway. Be- 
ginning at 0930, this regiment came ashore 
in a column of battalions with 2/1 in the 
tlie van followed by the 3d and 1st Bat- 
talions in that order. 

Artillery came next, and the units par- 
tially bogged down. The howitzer men 
admitted later that they had taken too 
much gear ashore with them. Prime mov- 
ers for the 105mm howitzers did not get 
ashore initially because there were not 
enough ramp boats for this work, and one- 
ton trucks proved too .light to handle the 

field pieces. Needed were two-and-a-half- 
ton six by sixes and ramp boats to put 
these vehicles on the beach simultaneously 
with the howitzers. Such prime movers 
were authorized, but so were a lot of other 
things the Marines did not have. 

In spite of these troubles, the artillery 
units reached their assigned firing posi- 
tions by making overland prime movers 
out of amphibian tractors that began to 
wallow ashore heavy with cargo.^ Once 
in position, however, the gunners found 
the amphibian was a creature of mixed 
virtues: tracked vehicles tore up the 
communications wire, creating early the 
pattern of combat events that became too 
familiar to plagued wiremen. 

Meanwhile the light 75mm pack howit- 
zers had made it ashore with little trouble, 
and the advance toward the airfield got 
underway. At 1115 the 1st Marines moved 
through the hasty perimeter of the 5th 
Marines and struck out southwest toward 
Mount Austen, the "Grassy Knoll." Cates 
put his regiment across the Tenaru at an 
engineer bridge supported by an amphib- 
ian tractor, and the 1st Marines pro- 
gressed slowly into the thickening jungle. 
Behind, to extend the beachhead, 1/5 
crossed the mouth of the Tenaru at 1330 
and moved toward the Ilu. Neither ad- 
vance encountered enemy resistance. 

' In early maps the names of the Tenaru and 
Ilu Rivers were incorrectly transposed. In this 
account the names will be applied to the correct 
rivers, but the name, "Battle of the Tenaru," 
will be retained to identify the August battle at 
the mouth of the Ilu. 

' The amphibian, later to be used to transport 
assault troops from ship to shore, started its war 
career here in a modest manner. Its "useful- 
ness exceeded all expectations,'" the Marines re- 
ported, but at the time nobody considered the 
strange craft capable of much more than amphib- 
ious drayage. The following day (D-plus one) 
on Gavutu an amphibian would create rather 
dramatic history by attacking a hostile cave, but 
such bravery was never recommended, even later 
when these craft entered the heyday of their more 
important role. 



Colonel Gates realized almost at once 
that it would be impossible to reach Mount 
Austen as his day's objective. The so- 
called Grassy Knoll, visible from the ships, 
could not be seen from the beach. It com- 
manded the Lunga area, but it lay much 
farther inland than reports of former 
planters and schooner pilots had indicated. 

Under heavy packs, sometimes excessive 
loads of ammunition, and with insufficient 
water and salt tablets,^ the 1st Marines by 
late afternoon had struggled but a mile 
when General Vandegrift ordered the 
regiment to halt, reorient, and establisli in- 
ternal contact. The men dug in a perim- 
eter in the jungle, some 3,500 yards south 
of the Ilu's mouth where 1/5 had ended 
its advance, to set up for the night. 

In spite of the breakneck pace with 
which the shoestring operation had 
mounted out and thrown itself in the path 
of the Japanese advance along the Solo- 
mon chain, the landing was a success. Al- 
though the lack of opposition (on the 
Guadalcanal side only) gave it somewhat 
the characteristics of a training maneuver, 
the need for additional training that Van- 
degrift had hoped to give his men in New 
Zealand became apparent. The general 
criticized the "uniform and lamentable" * 
failure of all units to patrol properly their 
fronts and flanks. 

Logistical difficulties were worse. Move- 
ment of supplies from the landing craft to 
the beaches and then to supply dumps soon 
began to snarl. Admiral Turner blamed 
this on the Marines' failure to understand 
the number of troops required for such 

^Medically speaking, the weight of individual 
equipment was excessive in most cases for men 
who had been cooped so long in steamy holds of 
ships and fed short, and sometimes inferior, 
rations. FinalRept, Phase II, 6. 

'Ibid., 10. 

work, failure to extend the beach limits 
promptly enough and, to some extent, a 
lack of control and direction over troops 
in the beach area. But the trouble and its 
causes were neither as clear-cut nor as 
damning as that. Marine planners had 
foreseen a dangerous shortage of man 
power at this critical point, but under the 
uncertain circumstances on this hostile 
beach they felt they could allot no more 
men to the job than the 500 from Colonel 
George R. Rowan's 1st Pioneer Battalion. 
Vandegrift did not want working parties 
to cut the strength of his fighting units to 
a level which might risk getting them de- 

Hindsight now makes it clear that the 
supplies mounting up as a juicy beach tar- 
get jeopardized the operation more than 
a call for additional working parties would 
have done. There were hardly enough 
Japanese fighting men ashore on the island 
to bother the Vandegrift force, but if en- 
emy planes from Rabaul had concentrated 
on hitting the congested beach they would 
have played havoc with this whole venture. 
Marines were aware of this risk, but they 
also expected to run into a sizable Jap- 
anese force somewhere in the thickening 
jungle. The people in the shore party 
would just have to work harder. 

Sailors joined the pioneers but the beach 
remained cluttered in spite of this help. 
Needed, division officers reported later, 
were "additional personnel in the propor- 
tion of at least 100 men for each vessel 
discharging cargo across the beach." ^ It 
was not that this problem had never been 
thought out and planned for in fleet exer- 
cises over the years. It was just that this 
was "Operation Shoestring." The situa- 
tion became so bad during the night of 7-8 

■ IMd., Phase V, 7. 



August that the landing force had to ask 
the ships to stop unloading. There had 
been air attacks that afternoon, and more 
were expected on the 8th. The exhausted 
workers needed time to clear the beaches 
and spread out the gear so it would be less 
of a target. 

Fortunately the air attacks during the 
day had concentrated on the shipping. At 
about 1100 on the Yth a coastwatcher in 
the Upper Solomons passed the word on 
the watchers' network that about 18 
bombers were on the way to Guadalcanal. 
This warning was relayed to Guadalcanal 
through Brisbane within 25 minutes, and 
the planes arrived at 1320. The destroyer 
Mugford suffered 20 casualties under a 
250-pound bomb hit, but it was the only 
ship struck by the attack. Antiaircraft 
fire downed two of the twin-engined Type 
97's. Later in the afternoon, at about 1500, 
10 Aichi dive bombers had no luck at all, 
but fire from the ships scratched another 
two Japanese planes. Other planes from 
both these attacks were downed by 
Fletcher's carrier aircraft. 

At 2200 on 7 August, Vandegrift is- 
sued his attack order for the following day. 
Plans had been changed. Since Mount 
Austen was out of reach, and because only 
10,000 troops were available in the Lunga 
area, he ordered an occupation of the air- 
field and establishment of a defensive line 
along the Lunga River. Positions east 
and southeast of Red Beach would be 
maintained temporarily to protect supplies 
and unloading until shore party activities 
could be established within the new 

At 0930 on 8 August the 1st Battalion, 
5th Marines and Company A, 1st Tank 
Battalion crossed the Ilu River at 
its mouth and advanced cautiously west- 

ward along the beach toward the Lunga. 
At the same time the 1st Marines moved 
from its night perimeter. Contact between 
units within this regiment was faulty, but 
by nightfall Lieutenant (^olonel Lenard 
B. (^resswell's 1st Battalion had overrun 
the field and reached the Lunga. The other 
two battalions, slowed by difficult terrain, 
advanced about 500 yards an hour and 
bivouacked for the night south of the 

Along the beach, 1/5 and the tanks met 
the first scattered resistance as they passed 
through the area in which the main Jap- 
anese force had been located. A few pris- 
oners were taken, and intelligence indi- 
cated that the enemy was in no position 
to attack the superior Marine landing 
force. Continued lack of resistance else- 
where seemed to confirm this, and at 1430 
the Marines contracted their front, crossed 
the Lunga by a bridge immediately north 
of the airfield, and advanced more rapidly 
toward the Kukum River, a stream in the 
western fan of the Lunga delta. 

With Company D leading, this advance 
came upon the main Japanese encampment 
area at 1500. The enemy force, obviously 
smaller than anticipated, had retreated in 
evident haste and confusion. Large quan- 
tities of undamaged food, ammunition, 
engineering material, electrical gear, and 
radio equipment had been left behind. Al- 
though some improperly indoctrinated 
Marines began to destroy this gear, that 
tendency soon was halted, and in the next 
few weeks these men would lose their con- 
tempt for this windfall of material. 

Except for token resistance from some 
of the straggling Japanese attempting to 
flee west, air action constituted the enemy's 
only effort to hamper the Marines. At 
about 1100 Coastwatcher Cecil John Ma- 



son, Pilot Officer, EAAF, warned from his 
Bougainville hide-out that a large num- 
ber of planes were winging toward Guadal- 
canal. In another hour some 40 twin- 
engine torpedo planes appeared over the 
area to find the task force, alerted by the 
warning, maneuvering at top speed while 
employing evasive tactics. 

A torpedo sent the destroyer Jarvis 
limping southeast for the New Hebrides. 
She was sunk next day by an enemy air 
attack. The transport Elliott, set afire 
when an enemy plane crashed aboard, had 
to be beached and destroyed by her sister 
ships. Survivors went on board the 
Hunter Liggett. 

Ship antiaircraft fire and fighter planes 
from Admiral Noyes' carriers shot down 
12 Japanese planes, and shore-based an- 
tiaircraft accounted for two more. Still 
others were splashed by carrier-based 
fighters west of the transport area. A 
total of seven American planes were lost.® 


These early attacks hampered Marine 
operations and unloading, but the beach- 
head continued to grow. The Japanese 
had no intentions of giving up their posi- 
tions in the Southern Solomons without 
a fierce fight, however, and early on 8 
August a task force of five heavy and two 
light cruisers and a destroyer made ready 
to strike American shipping in Sealark 

After rendezvousing at St. George's 
Channel off Rabaul, this force steamed 
south along Bougainville's east coast until 
it sighted an Allied patrol plane observing 
its course. Reversing, the ships made 

back up the island coast until the plane 
departed. Then turning again, they sailed 
between Bougainville and Choiseul north- 
east of the Shortlands and set course 
down "The Slot'" toward Guadalcanal. 

Word of this approaching force reached 
Admiral Turner at 1800, and, when Ad- 
miral Fletcher notified him shortly there- 
after that the carrier force was to be with- 
drawn, Turner called Vandegrift to the 
flagship McCawley and informed the gen- 
eral that, deprived of carrier protection, 
the transports must leave at 0600 the next 

As early as 2 August Admiral Ghormley 
had known of Fletcher's intentions to re- 
tire the carriers before D-Day plus three. ^ 
At 1807 on 8 August Fletcher cited fuel 
shortage and plane losses that had reduced 
his fighter craft from 99 to 78 and again 
requested permission to withdraw until 
sufficient land-based aircraft and fuel were 
available to support shipping.* It seems 
that Ghormley had not really expected 
this problem to come up, in spite of 
Fletcher's announcement about this mat- 
ter at the Fiji rehearsals. But now 
that Fletcher was making the request, 
Ghormley gave his approval. Ghormley 
explained later : 

When Fletcher, the man on the spot, informed 
me he had to withdraw for fuel, I approved. He 
knew the situation in detail ; I did not. This re- 
sulted in my directing Turner to withdraw his 
surface forces to prevent their destruction. I 
was without detailed information as to Turner's 
situation, but I knew that his forces had landed 
and that our major problem would become one 
of giving every support possible to Vandegrift." 

"OinOPac, "Preliminary Report, Solomon Is- 
lands Operation," August 1942, 4. 

'ComSoPac War Diary, 2Aug42 (located at 

" Ibid., 9Aug42. For a detailed discussion of 
Adm Fletcher's withdrawal of his carriers see 
also Struggle for Guadalcanal, 27-28, 117. 

" Ghormley MS, 93. 



Vandegrift held that retirement of the 
ships would leave him in a "most alarm- 
ing" position." Division plans assumed 
the ships would remain in the target area 
four days, and even then all available sup- 
plies would prove scanty enough, such was 
the haste with which the assault mounted 
out with less than the normal minimum in 
basic allowances. But a withdrawal early 
on 9 August would take much of the sup- 
plies and equipment away in the holds of 
ships and leave beach dumps in a state of 
chaos. The "shoestring" of this first Allied 
offensive seemed to be pulling apart. This 
was the first of the operation's many dark 
hours. (See Map 14, Map Section) 

While Vandegrift conferred with 
Turner, the Japanese ships, elements of 
the enemy's Eighth Fleets approached 
Savo Island undetected by destroyers 
Ralph Talbot and Blue on picket duty 
northwest of that small island. They 
slipped past these ships toward the two 
Allied cruisers, HMAS Canberra and the 
USS Chicago, and destroyers USS Bagley 
and Patterson which patrolled the waters 
between Savo and Cape Esperance. Far- 
ther north cruisers USS Vincennes, 
Astoria, and Quincy and destroyers Helm, 
and Wilson patrolled between Savo and 
Florida. Down the channel two cruisers 
with screening destroyers covered the 

With seaplanes up from his cruisers to 
scout for the Allied ships, Eighth Fleet 
Commander Rear Admiral Gunichi Mi- 
kawa steamed southeast until he sighted 
his enemy at about the same time Allied 
ships in Sealark Channel received reports 
of one or more unidentified planes. But 
Admiral Mikawa's surface force still was 
undetected at 2313 when these reports 

came in. Admiral Turner had estimated 
that the Japanese ships would hole up in 
Rekata Bay on Santa Isabel Island and 
strike at the amphibious force with tor- 
pedo-carrying floatplanes. 

At 0316 Mikawa ordered independent 
firing, and torpedoes leaped from their 
tubes two minutes later. Japanese float- 
planes illuminated briefly. The Canberra 
caught two torpedoes in her starboard side, 
the Chicago lost part of her bow, and then 
the Japanese turned toward the Allied 
ships between Savo and Florida. 

The resulting melee was one of the worst 
defeats ever suffered by the U. S. Navy. 
The YiTwennes and the Quincy were lost; 
the Australian Canberra burned all night 
and had to be abandoned and sunk; de- 
stroyer Ralph Talbot was damaged, and 
Astoria went down at noon the next day. 
Fortunately Mikawa retired without 
pressing his advantage in an attack on the 
amphibious shipping farther down the 
channel, and Admiral Turner, delaying 
his departure, ordered unloading to con- 
tinue. Late in the afternoon the trans- 
ports got underway for Noumea, leaving 
the Marines on their own with four units 
of fire and 37 days' supply of food. 

Even when loaded in Wellington the 
level of supplies and ammunition had been 
considered slim. That original loading 
of 60 days' supplies and 10 units of fire was 
respectively 33 and 50 per cent below the 
90-day and 20-unit levels then considered 
normal for operations of this kind." Now 
the ships had taken part of these loads 

' FinalRept, Phase II, 13. 

" Division staff officers admitted later that sup- 
plies for 60 days represented more gear than 
their slim fighting outfit could handle logisti- 
cally at the beach. They recommended that 
levels should be pegged at 30 days for general 
supplies, 50 days for rations. The lovcer level 



away, leaving a most inadequate fraction 
behind. And with air support so sketchy, 
there was no way to know when the trans- 
ports could come back again. The stacks 
of captured Japanese rations began to 
gain in importance if not in palatability. 
According to the war diary of the Com- 
mander, Task Force 62, the following 
troops were left in the Guadalcanal- 
Tulagi area when the transports and 
supply ships withdrew : 

At Guadalcanal: 

Division Headquarters Company (less de- 
tachments ) 

Division Signal Company (less detachments) 

5th Marines (less 2d Battalion) 

1st Marines 

11th Marines (less Battery E, 1st and 4th 

1st Tank Battalion (less detachments) 

1st Engineer Battalion (less detachments) 

1st Pioneer Battalion (less detachments) 

1st Amphibian Tractor Battalion (less de- 

1st Service Battalion (less detachments) 

1st Medical Battalion (less detachments) 

1st Military Police Company 

2d Platoon, 1st Scout Company 

Units, 3d Defense Battalion 

Local Naval Defense Force 

Total on Guadalcanal, aioiit 10,000 

At Tulagi: 

1st Raider Battalion 
1st Parachute Battalion 

2d Battalion, 5th Marines (2d Platoon, Com- 
pany A, 1st Pioneers attached) 
1st, 2d, and 3d Battalions, 2d Marines 
Batteries H and I, 3d Battalion, 10th Marines 
Detachment, Division Headquarters Company 
Detachment, 2d Signal Company 
3d Defense Battalion (less detachments) 
Company A, 1st Medical Battalion 
Company A, 2d Engineer Battalion (2d Pla- 
toon, Company A, 1st Engineer Battalion 
Company C, 2d Tank Battalion 

of 10 units of fire was just right, they added ; no 
attempt should be made to carry 20 units into 
future landings. FinalRept, Phase II, 17. 

Company A. 2d Amphibian Tractor Battalion 
(2d Platoon, Company A, 1st Amphibian 
Tractor Battalion attached) 

Company D. 2d Medical Battalion 

Company A, 2d Pioneer Battalion (2d Platoon, 
Company A, 1st Pioneer Battalion attached ) 

Battery E, 11th Marines 

Company C, 2d Service Battalion 

Local Naval Defense Force 

Total on Tulagi, 6,075 

Total personnel left in area, about 16,01o 

The 2d Marines under Colonel John M. 
Arthur had formed the division reserve 
and was originally slated for the occupa- 
tion of Ndeni, but all its battalions now 
were in action in the Tulagi area. The 
regimental headquarters remained afloat, 
however, as did working parties from all 
companies, most of the Headquarters and 
Service Company, Regimental Weapons 
Company, administrative units from the 
various battalions, and G and Headquar- 
ters and Service Batteries of 3/10. 

The sudden withdrawal of the trans- 
ports carried these units, which totaled 
about 1,400 officers and men, back to Es- 
piritu Santo where they were used to "re- 
inforce the garrison there," according to 
the reports of Admiral Turner. On 14 
August, Turner ordered Colonel Arthur to 
report for duty with the Commanding 
General, Espiritu Santo. But a few days 
later Colonel Arthur and a small number 
of his officers and men got back up to 

There seemed no question in Turner's 
mind about his unrestricted claim of "pos- 
session" of the Marines in his area. If 
his handling of Colonel Arthur was a 
rather ungenerous bypass of General 
Vandegrift's command territory, the ad- 
miral's plan for those Marines who re- 
mained at Espiritu Santo was an even 

' CTF 62 War Diary, September and October 




more glaring example of his theory of 
personal command possession. He ordered 
those "idle" 2d Marines to form a "2d 
Provisional Raider Battalion." Then he 
wrote to Admiral Ghormley recommend- 
ing an overhaul of all Marine regiments 
in the Amphibious Force, South Pacific. 
All regiments then would contain raider 
battalions which could be sent out on 
special missions. Turner said he did not 
think Marine regiments would be suited 
to operations in the Pacific. "The em- 
ployment of a division seems less likely," 
the admiral added. He would use raider 
battalions like building blocks, and fit the 
landing force to the special problem. Ob- 
viously, he expected the Pacific war to be 
small and tidy. 

Admiral Ghormley answered that 
Turner ought to hold up such reorganiza- 
tion until he found out what the Com- 
mandant of the Marine Corps thought of 
all this. Admiral Ghormley then sent 
this letter and his endorsement to the 
Commandant via Admiral Nimitz at Pearl 
Harbor. Nimitz agreed with Ghormley, 
and he stressed that "extemporized organ- 

ization of Marine Forces should be made 
only in case of dire necessity." Nimitz 
then forwarded this correspondence on to 
Lieutenant General Thomas Holcomb, 
Marine Commandant. 

General Holcomb responded to Nimitz 
that the latter's objections had surely 
stopped Turner's plan without the need for 
the Commandant to add other objections, 
but Holcomb noted "with regret" that 
Turner had not seen fit to ask General 
Vandegrift about this plan to reorganize 
his troops. 

This reaction from Nimitz, and the ar- 
rival at about that time in the New Heb- 
rides of the "authentic" 2d Raider Bat- 
talion of Lieutenant Colonel Evans F. 
Carlson, caused Turner to halt his plan to 
turn all Amphibious Force Marines into 
raiders. But it took the admiral much 
longer than this to abandon his theory that 
these Marines were direct "possessions" of 

" ComPhibForSoPac Itr to ComSoPacFor, 
29Aug42 ; ComSoPacFor Itr to CinCPOA, 6Sep42 ; 
CinCPOA Itr to CMC, 24Sep42 ; CMC Itr to 
CinCPOA, 30ct42. 


Tulagi and Gavutu-Tanambogo 


After Task Group Yoke separated from 
the larger body of ships at 0240 on D-Day, 
its approach to Tulagi was accomplished 
without incident. All elements of the 
group arrived in position at about 0630 ^ 
and made ready for the landing. 

As the ships approached the transport 
area, 15 fighters and 15 dive bombers from 
Wasp strafed and bombed the target area,^ 
setting fire to seaplanes that were caught 
in the harbor.' (See Map 15, Map Section) 

Five-inch naval gunfire from the de- 
stroyer Monssen^ opened up at a promon- 
tory of Florida Island, west of Tulagi, 
and 60 rounds were expended on the target 
between 0727 and 0732. In the meantime, 
both the Buchanan and San Juan (an 
antiaircraft cruiser) pumped 100 rounds 
each into nearby targets. Buchanan con- 
centrated on a point of land east of Haleta, 
on Florida, while the San Juan blasted 

' At 0625, Tulagi sent its message to Japanese 
stations to tlie north tliat an enemy surface 
force had entered the channel. Tulagi CommB 
msg of 7Aug42 in 25th AirFlot War Dairy, 
August-September 1942, hereinafter cited as 25th 
Air Plot Diary. 

' OomWaspAirGru Rept to CO Wasp, 10Aug42. 
In general, during the first day Wasp planes 
operated over the Tulagi area while Saratoga 
planes gave comparable support to the main 
landing off Beach Red at Guadalcanal. Enter- 
prise planes gave protection to the carriers and 
flew patrol missions. 

^ "0630 — All flying boats have been set aflre 
by the bombardment." CTF 18 ActRept, 
6-10Aug42, 1, hereinafter cited as CTF 18 AR. 

a small island south of the same point of 

At 0740, 20 minutes before H-hour, 
Company B (reinforced) of the 1st Bat- 
talion, 2d Marines, under command of 
Captain Edward J. Crane, landed on 
Florida near Haleta to protect the left 
flank of the Tulagi Force. The landing 
was unopposed, although enemy troops 
had been reported in position there on 25 
July.^ Crane, his company reinforced by 
the 4th platoon of Company D and 21 men 
from Headquarters Company, reached his 
objective within 40 minutes. The 252 offi- 
cers and men went ashore in eight landing 
boats and were guided to their objective 
by one of the several Australians on duty 
with the division." 

While this covering force deployed in- 
land from its Florida beach, the remainder 
of the 1st Battalion, 2d Marines (Lieu- 
tenant Colonel Robert E. Hill) made a 
similar security landing at Florida's 
Halavo Peninsula near Gavutu and 
Tanambogo. The craft drew some fire 
from Gavutu but there were no casualties, 
and no enemy forces were encountered on 
the peninsula. These Marines later re- 
turned to their ships. 

At Tulagi not a single landing craft of 
the first wave was able to set its passengers 
directly ashore. All of them, hung up on 

* Ibid., 2. 

'ComSoPac War Diary, 25Jul42 (located at 

' LtCol H. R. Thorpe Itr to CMC, 19Jan49. 




coral formations at distances varying from 
30 to well over 100 yards from the beach 
line, and the assault personnel of raider 
Companies B and D waded ashore against 
no opposition, through water initially 
from waist to armpit deep.^ 

Meanwhile the enemy defense forces, 
concentrated in the southeastern third of 
the island, realized that an all-out assault 
was underway. Between 0725 and 0749, 
the Tulagi Communication Base notified 
the Commanding Officer of the Ttventy- 
Fifth Air Flotilla at Rabaul that Tulagi 
was under bombardment, that the landings 
had begun, and that the senders were de- 
stroying all equipment immediately. At 
0800 the Japanese messages said shells were 
falling near the radio installation. Ten 
minutes later, the final message went out : 
"Enemy troop strength is overwhelming. 
We will defend to the last man." * 

Companies B and D had reached the 
beach, and the landing craft carrying 
raider Companies A and C now began to 
hang upon the coral. The Weapons Com- 
pany (Captain George W. Herring) of 
the raider battalion, whose 60mm mortars 
had been attached to the assault com- 
panies,® headed ashore to assume respon- 
sibility for beachhead security. 

Assaulting. Marines crossed the beach 
and moved up the face of a steep, heavily- 
wooded coral slope, the southwestern por- 
tion of the 350-foot ridge that forms an 
almost unbroken wall along the island's 
entire length. Major Lloyd Nickerson's 
Company B pushed on to the far coast of 
the island where it captured, without oppo- 

' Maj J. C. Erskine interview in HistDiv, 
HQMC, 15Mar49. 

' 25th AirFlot Diary. 

' Majs J. B. Sweeney, H. Stiff, W. E. Sperling 
interview in HistDiv, HQMC, 4Feb49, herein- 
after cited as Sweeney Interview. 

sition, the native village of Sasapi. This 
company then swung to the right and, ty- 
ing in with Major Justice Chambers' Com- 
pany D which had gained the high ground, 
began moving southeast. The advance of 
these two companies was steady and with- 
out opposition until Company B reached 
Carpenter's Wharf, halfway down the east 
shore of the island, where it encountered 
a series of enemy outposts. 

Meanwhile additional raiders had 
landed. Captain Lewis W. Walt's Com- 
pany A, landing to follow the leading 
companies, swung right atop the ridge 
spine, and tied in on the left with Com- 
pany D. Major Kenneth Bailey's Com- 
pany C also swung right, tied its left flank 
to Company A, and echeloned itself to the 
right rear to the beach. Spread out across 
the island, the raiders swept soutlieast 
against little opposition until Phase Line 
A, from the high ground northwest of Hill 
281 to Carpenter's Wharf, was reached 
at 1120. Here Major Chambers was 
wounded by mortar fire, and (\ptain 
William E. Sperling assumed command of 
Company D. 

By this time Colonel Edson, command- 
ing the 1st Raider Battalion, was ashore 
and ready to begin a coordinated attack to 
the southeast. Confronting him was the 
more thickly settled portion of the island 
where the British governmental activities 
had centered. This area is a saddle, be- 
tween the ridge first swept by the raiders 
and a smaller hill mass at the island's 
southeastern end." 

After directing a preparatory fire of in- 
fantry weapons into the area to their front, 

'" Tlie raiders liad been well briefed on the ter- 
rain of the island by Lt H. E. Josselyn, RANR, 
a former resident of the area who had intimate 
knowledge of it. Ibid. 



the raiders moved out toward the high 
ground beyond the saddle. Company C, 
on the right flank of the attack, drew fire 
almost immediately from Hill 208, a knob 
forward of the ridge that had just been 
cleared. The bulk of the Japanese resist- 
ance concentrated in the seaward face of 
the high ground, and Company C was 
caught by fire from enemy infantry 
weapons as it tried to pass between the 
hill and the beach. The raider company 
then turned its attack toward the hill and 
fought for nearly an hour before the Jap- 
anese positions were silenced. 

Radio communications between Edson 
and General Rupertus deteriorated rap- 
idly after this attack was launched, but the 
raider commander remained in contact 
with his fire support ships. Operation 
orders called for the various fire support 
sections to provide the landing force with 
naval gunfire liaison parties, and two of 
these were in Edson's CP with their 
radios.^* When the other raider companies 
came under fire from Hill 281 while Com- 
pany C fought against Hill 208, Edson put 
these naval gunfire teams to work. The 
San Juan fired a seven-minute, 280-round 
concentration of 6-inch shells onto Hill 
281. When it lifted the raiders advanced 
with a steady pressure against the enemy. 

Four hours later, at 1625, Edson notified 
Rupertus that 500 enemy had broken con- 
tact with his force and had withdrawn into 
the southeastern ridge. 

The advance continued slowly until 
dusk. At that time Company E (raiders) , 
relieved of the beach defense mission by 
2/5 which had landed at 0916, reported to 
its parent organization. Company D, now 
on the extreme left flank, had met little 

opposition since midmorning, when the 
first enemy encountered were flushed near 
Carpenter's Wharf by Company B. After 
this contact Company D pushed south 
along the eastern beach and at dusk 
reached the crest of Hill 281. Meanwhile 
Company B moved up again, now on the 
right of Company D, and gained high 
ground overlooking the cut of a cross- 
island roadway through the saddle be- 
tween Hills 281 and 230. Company D, on 
the far side of the road and to the left of 
B, took up night defensive positions with 
its right flank resting on the southern 
brink of the cut. Company B, augmented 
by elements of Headquarters Company, 
rested its left flank on the cut and extended 
its lines generally westward along the 
brink.12 Both companies put listening 
posts forward of the lines. 

Companies A and C (less one platoon) 
meanwhile encountered the terrain feature 
which harbored the island's most serious 
resistance. In the forward slope of Hill 
281, a deep ravine lay almost parallel to 
the raider advance and debouched several 
hundred yards southeast of Hill 208. Its 
sides were precipitous, and within it the 
enemy held strong positions which made 
assault hazardous. Maps which had been 
captured and translated during the day 
confirmed that this ravine would contain 
the core of enemy resistance. 

With further action against the pocket 
impossible at the time, all battalion ele- 
ments went into position for the night. 
Company E was placed on Company B's 
right, while Companies A and C (less one 
platoon) respectively tied in from the 
right of Company E. The positions ex- 
tended along high ground facing the ra- 

" VTF 18 AR, 2 ; Lt A. L. Moon Itr to LtCol B. D. 
Heinl, Jr., 13Feb49. 

' Sweeney Interview. 



vine's long axis, and listening posts were 

During Edson's sweep down the island, 
the 2d Battalion, 5th Marines (Rose- 
crans), had landed 1,085 officers and men 
and committed its units to various tasks. 
Company F scouted the northwest section 
of the island but met no opposition. At 
1000 Company E was ordered to operate 
generally in support of Company B (raid- 
ers), and one hour later the 3d Platoon of 
Company H (weapons) went forward to 
assist Company C (raiders) in the latter's 
attack against Hill 208. By 1300, when 
the raider battalion began its attack from 
Phase Line A, Company G moved down 
the trail along the ridge line and supported 
the raider battalion. Rosecrans' com- 
mand post later displaced southeast from 
near Beach Blue toward the scene of this 


The first night on Tulagi set the pattern 
for many future nights in the Pacific war. 
During darkness, four separate attacks 
struck the raider lines, and, although mi- 
nor penetrations occurred, the enemy made 
no attempt to consolidate or exploit his 
gains. The first attack, which met with 
some initial success, hit between Com- 
panies C and A. Outposts fell back to the 
main line of resistance (MLR), and the 
two companies were forced apart. The 
attack isolated Company C from the rest 
of the battalion, but the company was not 
molested again. Company A refused its 
right flank and awaited developments. 

They were not long in coming. Shift- 
ing the direction of his attack toward his 
right front, the enemy attempted to roll 

" Ihid. 

back Walt's men from the refused flank. 
But the flank held, killing 26 Japanese 
within 20 yards of the MLR. 

That ended the concerted attacks of the 
night. Thereafter, enemy efi^orts consisted 
entirely of attempts at quiet infiltration of 
the Marine positions. Individuals and 
small groups worked from the ravine 
through the raider lines and launched five 
separate small-scale attacks against the 
command post between 0030 and 0530. 
These were repulsed, and efforts on the 
part of two other enemy groups to skirt 
the beach flanks of Companies D and C 
likewise were turned back. 

On the morning of 8 August, two com- 
panies of the 2d Battalion, 5th Marines, 
moved up to assist in the sweep of the 
southeastern part of the island. Compa- 
nies E and F, 5th Marines, passed through 
Company D raiders, attacked down the 
forward slope of Hill 281, and swung right 
toward the enemy pocket in the ravine. 

Now flanking this troublesome terrain 
feature on three sides, Marines laid down a 
heavy mortar concentration from the 
60mm weapons of the raiders and 2/5's 
81s. By midafternoon the preparation 
was complete, and at 1500 the raiders and 
Company G, 5th Marines, pushed through 
the ravine to wipe out remaining resist- 
ance. This ended organized opposition on 
the island, and by nightfall of 8 August 
Tulagi was labeled secure. For several 
days, however, individual Japanese and 
small groups continued to be flushed from 
hiding places and hunted down by patrol- 
ling Marines. 


These islets, each dominated by a low, 
precipitous central hill of coral, are joined 



by a 500-yard causeway. Gavutu's hill, 
148 feet in height, stands some 25 to 30 feet 
higher than Tanambogo's highest point, 
and Gavutu thus became the main ob- 
jective of the landing which aimed at the 
higher ground. 

The plans " called for the landing to 
strike the northeast coast after an ap- 
proach from the east, and since Tanam- 
bogo lies approximately northwest of 
Gavutu the assault force faced the possi- 
bility of flanking fire from that island as 
well as frontal resistance from the main 
objective. Opposition from both islands 
was expected from the terrain dominating 
the flat beach. 

Naval gunfire and close air support by 
SBD's from the Wasp were expected to 
neutralize most enemy emplacements on 
these hills, but the fire plan did not reckon 
with the coral cave. Caves of this type 
began to appear as serious obstacles for 
the parachute battalion on Gavutu at about 
the same time the raiders began to en- 
counter them on Tulagi. 

Surprise was impossible. There were 
not sufficient craft for simultaneous land- 
ings, and the hour of assault was estab- 
lished in General Vandegrift's Operation 
Order Number 7-42 as H-plus four hours. 
So four hours after the raider landing on 
Tulagi, the parachute battalion made its 
frontal assault in the face of fire from an 
alerted garrison which was supported by 
fires from a flanking position. 

The battalion went ashore in three 
waves, one company per wave. The thor- 

"Ist Mar Div OpOrd No. 7-42, 20Jul42. See 
FinalRept, Phase II, Annex E, 2. Gavutu's im- 
portance stemmed from the islet's numerous in- 
stallations which included machine shops, jetties, 
and a radio station. USN ND Hydro, Vol. I — 
Sailing Directions For the Pacific Islands, 
(Washington: GPO, 1938, 4th ed.), 323. 

oughness with which the antiaircraft 
cruiser San Juan had carried out her fire 
support mission — 280 rounds of 6-inch fire 
against Gavutu in four minutes ^^ — and 
the intensity of the Wasp's dive-bombers" 
preparation caused heavy damage to the 
enemy installations, but this destruction 
actually worked to the disadvantage of 
the parachute battalion in one instance. 
The unit intended to land on a seaplane 
ramp from which the beach could be easily 
reached, but the ramp had been reduced 
to an unusable mass of rubble. Observ- 
ing this, the landing wave commanders al- 
tered course slightly to the north where 
craft became even more vulnerable to 
flanking fire. Part of the troops, scram- 
bling over a concrete pier that jutted four 
feet out of the water, were exposed to fire 
from both islands. General Vandegrift 
estimated that troops landing in this area 
suffered ten per cent casualties. 

Company A, the first wave, got ashore 
without casualties to work inland against 
no serious opposition. The four boats 
carrying Company B and the final wave, 
with Company C and miscellaneous at- 
tachments, came under fire as they neared 
the island. The landing succeeded, how- 
ever, and Company B, moving left and 
working toward Gavutu's southern end, 
gained some protection from enemy fire 
and continued to attack. 

Pinned down on the beach under heavy 
fire, the other companies made no advances 
until Company B gained high ground from 
which its fire assisted in getting the attack 
off the beach. Hill 148, Gavutu's high 
ground, was plastered by naval guns and 
assaulted on the east and southeast. By 
1430, Major Charles A. Miller, who had 
succeeded the wounded Major Robert H. 

CTF 18 AR, 2. 



Williams in command, controlled most of 
the island. Partially defiladed positions 
on Hill 148 's west-southwestern slopes, 
however, still were active, and enemy em- 
placements there and on Tanambogo 
threatened further advance. Miller re- 
quested reinforcements to complete the 
capture of both islands. 

In anticipation of their arrival. Miller 
also requested an air strike and naval gun- 
fire on Tanambogo, and Wasp planes fur- 
nished a 10-minute strike while Buchanan 
and Monssen, in position south of Gavutu, 
fired over that island and subjected the 
exposed faces of the hill on Tanambogo 
to an intense concentration of 5-inch shells. 

By this time all forces available to Gen- 
eral Rupertus had been committed, but 
since Captain Edward Crane's Company 
B (1/2) had met no opposition on Florida 
near Tulagi, this unit was ordered to re- 
port to Miller. The message reached the 
company just as landing craft arrived to 
withdraw the Marines from their Florida 

Embarked in six landing craft, the com- 
pany arrived at Gavutu at about 1800, and 
Miller directed Crane to land on Tanam- 
bogo and seize that island. Told that only 
a few snipers held the island, Crane guided 
his overcrowded craft around the east 
shore of Tanambogo according to direc- 
tions provided by Flight Lieutenant Spen- 
cer, RAAF, and under cover of darkness 
attempted a landing on a small pier on the 
northeastern tip of the island. (One boat, 
containing the 2d Platoon, hung up on a 
coral reef at Gavutu and took no part in 
the Tanambogo assault.) 

The first boat landed without incident, 
and the men deployed along the beach; 
but as the second boat discharged its men, 

' LtCol W. B. Kyle Itr to CMC, 10Feb49. 

a shell from one of the fire support ships 
ignited a nearby fuel dump, and the result- 
ing glare lighted the landing area and 
exposed the Marines. The enemy opened 
up immediately, taking all boats under 
rifle and machine-gun fire. Casualties 
mounted among the Marines ashore and 
still afloat, but the boat crews, being ex- 
posed, sufl^ered most heavily. One crew 
was completely wiped out and a Marine 
assumed control of the craft. 

The reinforcing machine-gun platoon 
(4th Platoon, Company D) in the second 
boat managed to set up two of its weapons 
on the pier, but intense enemy fire forced 
a withdrawal. 

In the meantime, Crane and about 30 
men had gone ashore. The intensity of 
resistance, however, made withdrawal in- 
evitable, and Crane succeeded in reembark- 
ing all wounded and all but 12 of the able 
survivors. The boats withdrew, some to 
Gavutu where they reported the event, and 
others direct to ships where the wounded 
were put aboard. Two of the men left 
ashore managed to return to Gavutu at 
about 2200 in a rowboat, while Crane and 
Lieutenant John J. Smith, leader of the 
2d Platoon, and the remainder of the dozen 
men made their way around the beach and 
over the causeway to arrive at Miller's 
Gavutu command post about midnight. 

At 2200, having been informed of the 
abortive attack on Tanambogo, General 
Rupertus requested the release of an addi- 
tional combat team. This request reached 
Vandegrift during his conference with 
Admiral Turner on board the USS Mc- 
Caivley, and Vandegrift, Turner concur- 
ring, released the remaining two battalions 
of the Division Reserve. At 0330, 8 
August, the USS President Hayes and 
President Adams, with the 1st and 3d Bat- 



talions, 2d Marines (reinforced) em- 
barked, were ordered to cross from the 
transport area off Guadalcanal's Beach 
Red to the Tulagi transport area. Simul- 
taneously battalion commanders received 
orders to land their troops at Beach 
Blue on Tulagi and report to General 

Upon arrival at the transport area off 
Beach Blue at 0730, the 3d Battalion was 
directed to pass to Gavutu, reinforce the 
troops engaged there, and seize Tanam- 
bogo. Orders for the 1st Battalion were 
cancelled and this unit did not land. 

The 3d Battalion, under Lieutenant 
Colonel Robert G. Hunt, landed on Gavutu 
in a succession of boat waves, with com- 
panies in the following order: Company 
L, with 5th Platoon, Company M attached, 
at 1000; Company K, with 4th Platoon, 
Company M attached, at 1025; Company 
I, with 3d Platoon, Company M attached, 
at 1050 ; Company M, less 3d, 4th, and 5th 
Platoons, with Headquarters Company, at 

Troops deployed initially to eliminate 
Gavutu opposition and to take Tanam- 
bogo under fire. Company L, for example, 
assumed positions generally around the 
base of Hill 148 facing Tanambogo, while 
Company K moved up the hill to relieve 
parachute b-attalion elements in positions 
there. At 1330 Company K had just 
accomplished its mission when an SBD 
pilot dropped a bomb within company 
positions on the northwest nose of the hill. 
Three men were killed and nine wounded. 

"CWO T. W. Huston Itr to CMC, 28Dec48. 
Orders to report to Rupertus did not go through 
Col J. M. Arthur, CO 2dMar. Each battalion 
commander was notified direct, and it was not 
until he reached Espiritu Santo that Arthur 
knew which of his troops had been committed. 
Col R. E. Hill interview at HistDiv, 18Apr49. 

Eight of the casualties were men of the 
supporting platoon of Company M. 

At 1225, Captain W. B. Tinsley, com- 
manding Company I, was ordered to pre- 
pare for a landing on Tanambogo. He 
would have the support of two tanks from 
Company C of the 2d Tank Battalion 
(one of the reinforcing units of the 2d 
Marines), and his attack would be pre- 
ceded by a 10-minute naval gunfire 
preparation by the Bxichanan. The com- 
pany would not be accompanied by its sup- 
porting machine-gun platoon, which was 
to stay in position on Gavutu and lay 
down supporting fires from there. 

At 1315 the tanks landed on Gavutu. 
Lieutenant E. J. Sweeney, commanding 
them, was ordered to land at 1615 on Tan- 
ambogo, using one tank to cover the south 
side of the hill on that island and the 
other to cover the eastern slope. 

The naval gunfire preparation began at 
1600. Twenty minutes later the assault 
company, following the tanks, made its 
landing. Lieutenant Sweeney was killed, 
but his tank rendered valuable support to 
the riflemen. The other tank, getting too 
far ahead of the assault troops, was dis- 
abled by an iron bar and set afire by oil- 
soaked rags employed by Japanese rifle- 
men. The entire enemy group was wiped 
out; 42 bodies were piled up around the 
disabled tank. 

At 1620 Company I landed and formed 
two attack groups. One worked up the 
southern slope of the Tanambogo hill 
while the other, moving to the right and 
then inland, attacked up the eastern slope. 
Japanese fought fiercely from caves and 
dugouts, and the eastern group drew fire 
from a few enemy riflemen and machine 
gunners on Gaomi, a tiny islet a few hun- 
dred yards east of Tanambogo. Naval 



gunfire from USS Gridley was directed 
upon Gaomi at 1700 and positions on the 
small island were silenced. At this time 
the 1st platoon of Company K attacked 
across the causeway from Gavutu, secured 
the Tanambogo end of the causeway, and 
took up positions for the night. 

By 2100, the southeastern two-thirds of 
the island had been secured, and at 2300 
a light machine-gun platoon from Com- 
pany M reported to Company I for sup- 
port against enemy counterattacks. Con- 
siderable close-in fighting took place dur- 
ing the night between the Marines and 
•Japanese who sallied from foxholes and 
dugouts. No change in position occurred, 
however, and by late the next day con- 
tinued attacks had secured the island. 

While Gavutu and Tanambogo were 
mopped up, the 1st and 2d Battalions, 2d 
Marines unloaded at Tulagi. The 1st 
Battalion, unengaged since its 7 August 
landing on Florida, went ashore at Beach 
Blue at 0900 on 9 August. The 2d Bat- 
talion (Major Orin K. Pressley) followed 
an hour later. 

Here, as at Guadalcanal, the amphibian 
tractor emerged as a versatile piece of 
equipment whose importance and utility 
could hardly be overestimated. From 
noon of 8 August throughout the follow- 
ing night, five of these vehicles of the 3d 
Platoon, Company A, 2d Amphibian 
Tractor Battalion (one of the reinforcing 
elements of the 2d Marines) operated be- 
tween Gavutu and the President Adams. 
They carried water, supplies, ammunition, 
and personnel to shore and evacuated 
wounded on the return trips. On one oc- 
casion a tractor moved some distance in- 
land to attack a Japanese position that had 
pinned down and wounded a number of 
Marines. Using their two machine guns. 

one .30 and one .50 caliber, the tractor's 
crew neutralized the enemy fire and then 
evacuated the wounded Marines.^® The 
five tractors of the platoon were taken 
back on board the Adams before sundown 
on 9 August. 

With the fall of Tanambogo, the last 
effective resistance in the Nggela island 
group ceased. Subsequent operations 
consisted of mopping up, consolidating 
defenses, and occupying several small 
peripheral islands including Makambo, 
Mbangai, Kokomtumbu, and Songonan- 

The mission of clearing out these small 
islands fell to various units of the 2d 
Battalion, 2d Marines. Makambo was 
taken by Company E, Mbangai by Com- 
pany F, and Kokomtambu and Songonan- 
gona, by Company G. Occupation of all 
these smaller islands was completed 
during the morning of 9 August. In all 
cases, opposition was slight. 

Occupation of the entire island group 
and destruction of the Japanese garrison 
had been accomplished in three days. The 
few prisoners taken were questioned and 
sent to rear areas. Most of them finally 
were placed in , a prisoner of war camp 
near Featherstone, New Zealand. 

Comparatively, the American losses 
were not excessive. An early report by 
Rupertus to the effect that the parachute 

" ". . . this was an emergency undertaking 
only as it is not considered that the tractor is 
a tactical combat vehicle." Final Rept, Phase II, 

" Spelling of place names are those which 
appear in Sailing Directions for the Pacific Is- 
lands, op. cit. The versions given there differ in 
numerous cases from those used in official reports 
of the campaign. Kokomtambu, for instance, 
appears in at least three different guises, while 
Songonangona surrendered its musical name to 
emerge as "Singsong" Island. 



battalion had suffered 50-60 per cent 
casualties can only be explained in terms 
of inadequate communications between 
him and his troops ashore. 

The exact number of Japanese casualties 
will never be known. An estimated 750- 
800 enemy were present in the Tulagi- 
Gavutu-Tanambogo area at the time of 
the landings. Twenty-three prisoners 
were taken, and an intelligence summary 
gives 70 as the approximate number of 
survivors who escaped to Florida. 

Immediately after organized resistance 
ceased and the isolated defending groups 
were rounded up or wiped out, Tulagi and 
its satellite islands were organized for de- 
fense against counterattack. The 1st 
Parachute Battalion, depleted by its ex- 
perience on Gavutu, moved from that 
island at 1700 on 9 August to Tulagi, where 
it went into position in the Government 
building area. The 2d Battalion, 5th 
Marines occupied the southeastern sector 
of the island, while two battalions of the 
2d Marines took over the defensive mis- 
sion in the northwest. The 1st Battalion 
occupied the extreme end of the island 
while the 2d Battalion established posi- 
tions at Sasapi. Third Battalion, 2d 
Marines, took over the occupation and 
defense of Gavutu, Tanambogo, and 

The logistic problem on Tulagi was a 
miniature of that encountered on Guadal- 
canal, although certain details were pe- 
culiar to Tulagi. The beachhead, for 
instance, was severely restricted by the 
abrupt ridge and, there were no usable 

"Col C. p. Van Ness Itr to CMC, 12Jan49. 
Defense initially was oriented against an antici- 
pated attack from Fldrida and artillery positions 
were selected with this, as well as the possibility 
of a sea-borne attack, in view. LtCol M. L. 
Curry interview at HistDiv, 2{?.Tan49. 

roads. Only after noon of the second day 
was it possible to move supplies ashore at 
the piers on the eastern coast. Both 
Gavutu and Tanambogo were so small that 
only ammunition and water were landed 
until the islands were secured. 

Naval gunfire on this side of the Solo- 
mon Islands operation had more of a work- 
out than it had received across the channel 
at Guadalcanal where opposition was at 
first light, but it was not an unqualified 
success. As a matter of fact it was "very 
poor," according to naval headquarters in 
Washington.^^ But this failing was caused 
mostly by lack of intelligence and time for 
planning and coordinated training. Im- 
proper ordnance made for another failing. 
Only armor-piercing shells could have 
blasted the Japanese from their caves, but 
the ships repeatedly fired high-capacity 
bombardment projectiles. Although many 
Naval officers were still of the opinion that 
a ship was a "fool to fight a fort," some 
began to agree with the Marine Corps that 
naval gunfire properly employed could be 
a big help in an amphibious assault. It 
was a case of the gunfire ships needing to 
move in closer for their fire missions. The 
commander of one ship reported : 

It was observed that the enemy had not been 
driven from the beach at Gavutu by the shelling 
and bombing preceding the landing. Further- 
more Tanambogo withstood two days of inter- 
mittent bombing and straffing and was not taken 
until a destroyer closed in to point blank range 
and shelled it for several minutes. It was evi- 
dent that this fire was necessary to Insure the 
capture of Tanambogo without further heavy 

Taking into account the indications that 
these shortcomings would be corrected in 

" CominCh, "Battle Experiences, Solomons Is- 
land Action," Information BuUetinNo. 2 (located 
atNHD),ChapX, 10. 

" USS Heyu'ood Kept, 12Aug42, 3. 



TULAGI 1S1-,AND , framed against the background of the larger Florida Island, is f re-swept from the hits 
scored by American carrier dive bombers. ( USN 11649) 

TANAMBOGO AND GAVUTU ISLANDS photographed immediately after a pre-landing strike by 
USS Enterprise planes; Gavutu is at the left across the causeway. (USN 11034) 



later operations, the Marine Corps was 
generally satisfied with the ships' fire. 
"The operation did not involve a real 
test . . . [but] nothing developed dur- 
ing the operation to indicate the need for 
any fundamental change in doctrine." " 

After these three days of fighting in the 
Tulagi area, this side of the operation re- 
mained quiet. Enemy planes bypassed it 
to strike at the more tempting Guadal- 

" Final Kept, Phase V, 6. 

canal airfield and perimeter. Surface 
craft shelled Tulagi occasionally, but never 
was it subjected to the kind of bombard- 
ment that struck Guadalcanal in October. 
There is no record that enemy reinforce- 
ments landed either on Tulagi or on Flor- 
ida Island. With this sharp fighting out 
of the way, the division could give all its 
attention to things on the larger island of 
Guadalcanal. There the picture was not 
a bright one. 


The Battle of the Tenaru 

With naval support gone, about the only 
hope was the airfield. Shipping would 
need air cover before regular runs could 
bolster the Marines' slim supply levels, and 
time was of the essence. If the Japanese 
struck hard while the landing force was 
abandoned and without air support, the 
precarious first step toward Rabaul might 
well have to be taken all over again. Van- 
degrift centered his defense at the field 
and gave completion of the strip top pri- 
ority equal to the task of building the per- 
imeter's MLR. 

On 8 August, almost as soon as the field 
was captured. Lieutenant Colonel Frank 
Geraci, the Division Engineer Officer, and 
Major Kenneth H. Weir, Division Air 
Officer, had made an inspection of the Jap- 
anese project and estimated the work still 
needed. They told Admiral Turner that 
2,600 feet of the strip would be ready in 
two days, that the remaining 1,178 feet 
would be operational in about two weeks. 
Turner said he would have aircraft sent in 
on 11 August. But the engineer officer 
had made his estimate before the trans- 
ports took off witli his bulldozers, power 
shovels, and dump trucks.^ 

Again, however, the Marines gained 
from the Japanese failure to destroy their 
equipment before fleeing into the jungle. 

' "The failure to land engineer equipment and 
machinery severely handicapped our efforts to 
complete the airfield and its defenses. Construc- 
tion equipment and personnel are not a luxury 
but an absolute necessity in modern warfare." 
Final Rcpt, Phase III, 12-13. 

Already the U. S. forces were indebted to 
the enemy for part of their daily two 
meals, and now they would finish the air- 
field largely through the use of enemy 
tools. This equipment included nine road 
rollers (only six of which would work), 
two gas locomotives with hopper cars on a 
narrow-gauge railroad, six small generat- 
ors (two were damaged beyond repair), 
one winch with a gasoline engine, about 50 
hand carts for dirt, some 75 hand shovels, 
and 280 pieces of explosives. 

In spite of this unintentional assistance 
from the Japanese, the Marine engineers 
did not waste any affection on the previous 
owners of the equipment. The machinery 
evidently had been used continuously for 
some time with no thought of mainte- 
nance. Keeping it running proved almost 
as big a job as finishing the airfield, and 
one of the tasks had to be done practically 
by hand, anyway. The Japanese had 
started at each end of the airstrip to work 
toward the middle, and the landing had 
interrupted these efforts some 180 to 200 
feet short of a meeting. Assisted by a few 
trucks and the narrow gauge hopper cars 
(which had to be loaded by hand), engi- 
neers, pioneers, and others who could be 
spared moved some 100,000 cubic feet of 
fill and spread it on this low spot at mid- 
field. A steel girder the Japanese had in- 
tended to use in a hangar served as a drag, 
and a Japanese road roller flattened and 
packed the fill after it had been spread. 




Problems facing the infantry troops 
were just as great. There had been no im- 
pressive ground action on Guadalcanal 
since the landing, but intelligence in the 
immediate vicinity as well as in the South 
Pacific in general was not yet able to indi- 
cate when, how, and where Japanese re- 
action would strike. Estimating a 
counterlanding to be the most probable 
course of Japanese action. General Vande- 
grift placed his MLR at the beach. There 
the Marines built a 9,600-yard defense 
from the mouth of the Ilu River west 
around Lunga Point to the village of 
Kukum. The Ilu flank was refused 600 
yards inland on the river "s west bank, and 
at Kukum the left flank turned inland 
across the flat land between the beach and 
the first high ground of the coastal hills. 
The oth Marines (less one battalion) held 
the left sector of the line from Kukum to 
the west bank of the Lunga, and the re- 
mainder of the line (inclusive of the 
Lunga) was held by the 1st Marines. (See 
Map 16) 

The line was thin. The bulk of the com- 
bat forces remained in assembly areas 
inland as a ready reserve to check attacks 
or penetrations from any sector. Inland 
(south) of the airfield, a 9,000-yard stretch 
of rugged jungle terrain was outposted 
by men from the artillery, pioneer, engi- 
neer, and amphibian tractor battalions. 
These men worked during the day and 
stood watch on the lines at night. 

The workers on the airfield as well as 
those on the thin perimeter were under 
almost constant enemy observation. Sub- 
marines and destroyers shelled the area at 
will day or night. Large flights of high- 
level bombers attacked the airfield daily, 
and observation planes were continually 
intruding with light bombs and strafing 

attacks. At night the enemy patrols be- 
came increasingly bold, and troops on the 
MLR mounted a continuous alert during 
the hours of darkness. South of the air- 
field the outpost line had to be supple- 
mented by roving patrols. 

In spite of this harassment, the perim- 
eter shaped up. The 1st Special Weap- 
ons Battalion dug in its 75mm tank de- 
stroyers (half-tracks) in positions inland 
from the beach, but kept them ready to 
move into prepared positions near the 
water. Howitzers of the 11th Marines 
were situated to deliver fire in all sectors. 
The 2d and 3d Battalions of the artillery 
regiment had 75mm pack howitzers and 
the 5th Battalion had 105mm howitzers. 
There were no 155mm howitzers or guns 
for counterbattery, there was no sound- 
flash equipment for the location of enemy 
batteries, and the 3d Defense Battalion 
had not had a chance to unload its 5-inch 
seacoast guns or radar units prior to 
the departure of the amphibious shipping. 
Air defense within the perimeter also was 
inadequate. There were 90mm antiair- 
craft guns ashore, but the restricted size 
of the perimeter kept them too close to the 
field for best employment. 

It was a hazardous and remote toe-hold 
which the Marines occupied, and within 
the Pacific high command there were some 
grave doubts whether they could hang on. 
Major General Millard F. Harmon said to 
General Marshall in a letter on 11 August : 

The thing that impresses me more than any- 
thing else in connection with the Solomon action 
is that we are not prepared to follow up. . . . 
We have seized a strategic position from which 
future operations in the Bismarcks can be 
strongly supported. Can the Marines hold it? 
There is considerable room for doubt.^ 

= CGSoPac Itr to CofSA, llAug42 (located at 



Admiral Ghormley, also concerned 
about the precarious Marine position,^ on 
12 August ordered Admiral McCain's TF 
63 to employ all available transport ship- 
ping to take aviation gasoline, lubricants, 
ammunition, bombs, and ground crews to 
Guadalcanal. To avoid Japanese air at- 
tacks, the ships were to leave Espiritu 
Santo in time to reach Sealark Channel 
late in the day, unload under cover of 
darkness, and depart early the following 

For the "blockade run" to Guadalcanal, 
Admiral McCain readied four destroyer 
transports of TransDiv 12. They were 
loaded with 400 drums of aviation gaso- 
line, 32 drums of lubricant, 282 bombs 
from 100- to 500-pounders, belted ammu- 
nition, tools, and spare parts. Also on 
board were five officers and 118 Navy en- 
listed men from a Navy construction base 
(Seabee) unit, Cub-1. Under the com- 
mand of Major C. H. Hayes, executive 
officer of VMO-251, this unit was to aid 
the Marine engineers in work at the field 
and to serve as ground crews for fighters 
and dive bombers scheduled to arrive 
within a few days. 

McCain's ships arrived off Guadalcanal 
during the night of 15 August, and the 
equipment and men were taken ashore. 
By this time the Marine engineers had 
filled the gap in the center of the landing 
strip and now labored to increase the 
length of the field from 2,600 feet to nearly 
4,000 feet. Work quickened after the Sea- 
bees landed, but there was no steel matting 
and the field's surface turned to sticky 

' He warned Adms King and Nimitz that Guad- 
alcanal might again fall to the Japanese if carrier 
support and reinforcements were not made avail- 
able. ComSoPac msgs to CinCPac and CominCh, 
16 and 17Aug42, in SoPac War Diary (located at 

mud after each of the frequent tropical 

General Harmon blamed a faulty plan- 
ning concept for the serious shortages of 
tools, equipment, and supplies. The cam- 
paign, he said, ". . . had been viewed by 
its planners as [an] amphibious operation 
supported by air, not as a means of estab- 
lishing strong land based air action." * 

But in spite of these shortages at the 
airfield and elsewhere, the Lunga Point 
perimeter was taking on an orderly routine 
of improvement and defense. Motor 
transport personnel had put their meager 
pool of trucks into operation shortly after 
the landing, and they had added some 35 
Japanese trucks to the available list. 
Pioneers had built a road from the airfield 
to the Lunga River where they erected a 
bridge to the far side of the perimeter. 
Supply dumps also had been put in order. 
The pioneers cleared the landing beach, 
moving gear west to the Lunga-Kukum 
area, and sorted and moved Japanese sup- 
plies. The old Japanese beach at Kukum 
was cleaned up and reconditioned to re- 
ceive U. S. material. 

Most of the work of moving the beach 
dumps to permanent sites was completed 
in four days. There was a great amount 
of tonnage to handle in spite of the fact 
that only a small portion of the supplies 
had been landed. Amphibian tractors and 
all available trucks, including the Japa- 
nese, were used. The Government Track 
(the coastal road to Lunga) was improved 
and streams and rivers bridged to speed 
truck traffic. The amphibians carried 
their loads just offshore through shallow 
surf, and farther out to sea the lighters 
moved from old beach to new and back 

'CGSoPac Itr to CGAAF, 23Aug42 (located at 



again. The amount of supplies at each 
of the new classified dumps was kept low 
to avoid excessive loss from bombardment. 
Captured material included almost 
every type item used by a military force — 
arms, ammunition, equipment, food, cloth- 
ing, fuel, tools, and building materials : 

As the division was acutely short of every- 
thing needed for its operation, the captured ma- 
terial represented an important if unforeseen 
factor in the development of the airfield and 
beach defenses and the subsistence of the gar- 

The landing force was particularly short 
of fuel, but in this case the supply left 
behind by the Japanese garrison was not 
as helpful as it might have been. Marines 
found some 800 to 900 drums of Japanese 
aviation gasoline on Guadalcanal, but this 
90-octane fuel was not quite good enough 
for our aircraft, and it was too "hot" and 
produced too much carbon in trucks and 
Higgins boats unless mixed evenly with 
U. S. 72-octane motor fuel. Likewise some 
160,000 gallons of Japanese motor fuel of 
60 or 65 octane proved unreliable in our 
vehicles although some of it was mixed 
with our fuel and used in emergency in 
noncombat vehicles. 

So critical was the supply of gasoline 
and diesel fuel that the division soon 
adopted an elaborate routine of "official 
scrounging" from ships that came into the 
channel. Rows of drums were lined bung 
up on old artillery lighters, and these craft 
would wallow alongside ships where Ma- 
rines would ask that a hose from the ships' 
bulk stores be passed over so they could 
fill the drums one at a time. This method 
helped the Marines' fuel supply, but not 
relations with the Navy. Small boats tak- 
ing off supplies had difficulty negotiating 

" FitmlRept, Phase III, 4. 

their passes alongside with the unwieldy 
lighters in the way, and ships officers quite 
frequently took a dim view of dragging 
along such bulky parasites when they had 
to take evasive action during the sudden 
air raids. But the system often worked 
well when early preparations were made 
with particularly friendly ships. 

This over-water work in Sealark Chan- 
nel, maintaining contact between Tulagi 
and Guadalcanal as well as meeting the 
supply ships which began to sneak in more 
frequently, pointed to another serious de- 
ficiency : there was no organized boat pool 
available to the division. More often than 
not the personnel and craft that the divi- 
sion used in those early days had merely 
been abandoned when the attack force de- 
parted, and there was no semblance of 
organization among them. Even the 
creation of order did not solve all the 
problems. A high percentage of the boats 
were damaged and crewmen had no repair 
facilities. The situation was gradually 
improved but was never satisfactory. 

At last, on 18 August, the engineers and 
Seabees had a chance to stand back and 
admire their work. The airfield was com- 
pleted. On 12 August it had been declared 
fit for fighters and dive bombers, but none 
were immediately available to send up. A 
Navy PBY had landed briefly on the strip 
on that date, but this was before Admiral 
McCain made his initial blockade run, and 
there was very little fuel for other planes 
anyway. But by the 18th the fill in the 
middle had been well packed, a grove of 
banyan trees at the end of the strip had 
been blasted away to make the approach 
less steep, and newly-arrived gasoline and 
ordnance were ready and waiting for the 
first customers. In the South Pacific 
during that period of shoestring existence. 



MARINE COMMANDERS ON GUADALCANAL appear m a picture taken four days after the land- 
ing which includes almost all the senior officers who led the 1st Marine Division ashore. Left to right, front 
row; Col George R. Rowan, Col Pedro A. del Valle, Col William C. James, MajGen Alexander A. Vande- 
grift. Col Gerald C. Thomas, Col Clifton B. Gates, LtCol Randolph McC. Pale, and Cdr Warwick T. 
Brown, USM. Second row: Col William J. Whaling, Col Frank B. Goettge, Col LeRoy P. Hunt, LtCol 
Frederick C. Biebush, LtCol Edwin A. Pollock, LtCol Edmund J. Buckley, LtCol Walter W. Barr, and 
LtCol Raymond P. Coffman. Third row: LtCol Francis R. Geraci, LtCol William E. Maxwell, Col Ed- 
ward G. Hagen, LtCol William N. McKelvy, LtCol Julian N. Frisbee, Maj Milton V. O'Connell, Maj 
William Chalfont, III, Capt Horace W. Fuller, and Maj Forest C. Thompson. Fourth row: Maj Robert 
G. Ballance, Maj Henry W. Buse, Jr., Maj James G. Frazer, Maj Henry H. Crockett, LtCol Leonard B. 
Cresswell, LtCol John A. Bemis, Maj Robert B. Luckey, LtCol Samuel G. Taxis, and LtCol Eugene H. 
Price. Last row: Maj Robert O. Bowen, LtCol Merrill B. Twining, Maj Kenneth W. Benner (behind 
Bemis), LtCol Walker A. Reeves, LtCol John DeW. Macklin, LtCol Hawley C. Waterman, and Maj James 
C. Murray. Many of these men went on to become general officers and three of them ( Vandegrift , Gates, and 
Pate) later became Commandants of the Marine Corps. ( USMC 50509 ) 



however, "readiness" was a comparative 
thing. There were no bomb handling 
trucks, carts or hoists, no gas trucks, and 
no power pumps. 

The state of readiness had a way of 
fluctuating rapidly, too, and the breathing 
spell for the workers did not last long. 
With most sadistic timing, a large flight 
of Japanese planes came over and scored 
17 hits on the runway. One engineer was 
killed, nine were wounded, and the field 
"was a mess." * 

The runway damage was disquieting but 
not altogether a surprise. Air raids had 
been frequent, shelling from offshore sub- 
marines likewise was common, and planes 
droned overhead frequently during the 
hours of darkness to drop small bombs 
here and there at well-spaced intervals. 
After the big raid on the 18th, the well- 
practiced repair teams merely went to 
work again. In filling the craters, the 
engineers and Seabees first squared the 
holes with hand shovels and then air ham- 
mered the new dirt solid by tamping every 
foot and a half of fill. They had found 
that this system kept settling to a mini- 
mum and prevented dangerous pot holes. 

Two days later Henderson Field, named 
after Major Lofton Henderson, a Marine 
aviator killed at Midway, was again 
ready. And this time the planes arrived. 
The forward echelon of Marine Aircraft 
Group 23, the first arrivals, numbered 
19 F4F"s and 12 SBD-3's. The units were 
under the command of Lieutenant Colonel 
Charles L. Fike, executive officer of the 
air group. The F4F's, a part of Marine 
Fighter Squadron 223, were commanded 
by Major John L. Smith, and the 12 
Douglas dive bombers from Marine Scout- 

'/6trf., Annexe, 2. 

Bomber Squadron 232 were led by Lieu- 
tenant Colonel Richard D. Mangrum. 

Arrival of planes ended an era for the 
Guadalcanal defenders — the hazardous 
period from 9 to 20 August when the land- 
ing force operated entirely without air or 
surface support. During this period lines 
of communications were most uncertain. 
Nothing was known of the general naval 
situation or the extent of losses at sea, and 
little information was received from aerial 
reconnaissance from rear areas. Ashore, 
patrolling was constant, but the terrain 
was such that much could be missed. Short 
rations, continuous hard work, and lack 
of sleep reflected in the physical condition 
of the troops. Morale, however, remained 

Formed in March of 1942 at Ewa, Oahu, 
MAG-23 remained in training there, with 
much shifting of personnel and units, until 
this two-squadron forward echelon sailed 
to the South Pacific on 2 August on board 
the escort carrier Long Island. Smith's 
men had just been issued new F4F's with 
two-stage superchargers, and Mangrum's 
unit had turned in its old SBD-2's for the 
newer 3's with self-sealing gasoline tanks 
and armor plate. The remaining two 
squadrons of the group, Captain Robert E. 
Galer's VMF-224 and Major Leo R. 
Smith's VMSB-231, would sail from the 
Hawaiian area on 15 August. 

John Smith's VMF-223 and Mangrum's 
VMSB-232 came down by way of Suva in 
the Fijis and Efate in the New Hebrides. 
At Efate, Smith traded some of his young, 
less-experienced pilots to Major Harold 
W. Bauer's VMF-212 for some fliers with 
more experience. On the afternoon of 20 
August, the Long Island stood 200 miles 
southeast of Guadalcanal and launched 
the planes. 



Two days later, on 22 August, the first 
Army Air Force planes, five P-400's ^ of 
the 67th Fighter Squadron, landed on the 
island. On 24 August, 11 Navy dive bomb- 
ers from the battle-damaged Enterprise 
moved to Henderson Field to operate for 
three months, and on 27 August nine more 
Army P-400's came in. Performance of 
these Army planes was disappointing. 
Their ceiling was 12,000 feet because they 
had no equipment for the British high- 
pressure oxygen system with which they 
were fitted, and they could not reach the 
high-flying enemy planes. Along with the 
Marine SBD's, the P-400's spent their time 
during Japanese air raids off strafing and 
bombing ground targets, and they returned 
to Henderson after the hostile planes de- 

A short while later — early in Septem- 
ber — supply and evacuation flights were 
initiated by two-engined Il4D's (C-47's) 
of Marine Aircraft Group 25. Flying 
daily from Espiritu Santo and Efate, the 
cargo planes each brought in some 3,000 
pounds of supplies and were capable of 
evacuating 16 stretcher patients. 

Although this increased air activity at 
Henderson Field was of great importance 
to the operation in general and the combat 
Marines in particular, the field still was 
not capable of supporting bombers which 
could carry attacks to Japanese positions 
farther to the north. On 20 August Gen- 
eral Harmon voiced the opinion that it 
would be too risky to base B-l7's at Hen- 
derson until more fighter and antiaircraft 
protection were available. ^ 

' Early P-39 "klunkers" converted for export 
to the British. They could carry one bomb, were 
armed with a 20mm cannon, two .50 caliber, and 
four .30 caliber machine guns. 

^CGSoPac Summary of Situation, 20Aug42 
(located at NHD). 

Early in September he suggested that 
heavy bombers stage through the Guadal- 
canal field from the New Hebrides and 
thus strike Rabaul and other Japanese 
bases,* but a closer investigation pointed 
up the impracticality of this plan. It 
would have meant hand-pumping more 
than 3,500 gallons of gasoline into each 
bomber landing at Guadalcanal on the 
1,800-mile round trip from the New 
Hebrides to the Northern Solomons; and 
although this manual labor was not too 
great a price to pay for an opportunity 
to strike at the Japanese, it was impossible 
to maintain a fuel stock of that propor- 
tion at Henderson Field. 


Combat troops meanwhile probed the 
jungles with patrols, and early reconnais- 
sance indicated that the bulk of Japanese 
troops was somewhere between the Matan- 
ikau River, about 7,000 yards west of the 
Lunga, and Kokumbona, a native village 
some 7,500 yards west of the Matanikau. 
General Vandegrift wanted to pursue the 
enemy and destroy him before the Japan- 
ese could reinforce this small, disorganized 
garrison, but no substantial force could be 
spared from the work of building the MLR 
and the airfield. 

Minor patrol clashes occurred almost 
daily, but many of these meeting engage- 
ments were with wandering bands of uni- 
formed laborers who only confused at- 
tempts to locate the main enemy force. 
This patrolling gradually revealed that 
the area between the Matanikau and Ko- 
kumbona was the main stronghold, how- 
ever. Stiff resistance continued there with 
each attempt to probe the area, and this 

"CGSoPac Itr to CofSA, 9Sep42 (located at 



pattern had started as early as 9 August 
when one officer and several enlisted men 
were wounded while trying to cross the 
river. This patrol had reported the west 
bank of the river well organized for de- 
fense. The enemy kept shifting his posi- 
tion, though, to maneuver for an 
advantage against the patrols which came 
to seek him out. Final confirmation of 
the enemy location came on 12 August 
when a Japanese warrant officer captured 
behind the 1st Battalion, 5th Marines said 
that his unit was between the Matanikau 
and Kokumbona. 

Under questioning, the prisoner admit- 
ted that possibly some of his fellow garri- 
son mates were wandering aimlessly 
through the jungle without food and that 
some of them might surrender. First 
Sergeant Stephen A. Custer of the division 
intelligence section made plans to lead an 
amphibious patrol to the area. Mean- 
while, a Marine patrol reported seeing 
what it took to be a white flag west of the 
river. Hearing these reports. Lieutenant 
Colonel Frank Goettge, division intelli- 
gence officer, decided to lead the patrol 

The original plan had called for an early 
start so that a daylight landing could be 
made. The patrol then would work inland 
along the west bank of the Matanikau and 
bivouac for the night far back in the hills. 
The second day was to be spent in a cross- 
country return to the perimeter. The pri- 
mary mission of the patrol would be that 
of reconnaissance, but it was to be strong 
enough for combat if it ran into a fight. 

Colonel Goettge's new plans delayed de- 
parture of the patrol about 12 hours and 
cut down its combat potential by including 
among its 25-man strength Lieutenant 
Commander Malcolm L. Pratt, assistant 

division surgeon, Lieutenant Ralph Cory, 
a Japanese linguist, and several members 
of the 5th Marines intelligence section. 
The boat got away from the perimeter at 
about 1800 and landed after dark at 2200 
at an undetermined point west of the 
Matanikau. The Japanese, instead of sur- 
rendering, attacked the patrol and cut off 
from the beach all but three men who 
escaped back into the surf to swim and 
wade to safety. 

One of these men. Sergeant Charles C. 
Arndt, arrived in the perimeter at about 
0530 on 13 August to report that the patrol 
had encountered enemy resistance. Com- 
pany A, 5th Marines set off immediately 
as a relief patrol to be reinforced later by 
two platoons of Company L and a light 
machine-gun section. Meanwhile, the 
other two escaped patrol members, Corpo- 
ral Joseph Spaulding and Platoon Ser- 
geant Frank L. Few, came back at 0725 
and 0800 respectively and revealed that the 
remainder of the Goettge patrol had been 
wiped out. 

The relief patrol landed west of Point 
Cruz, a coastal projection a short distance 
west of the Matanikau's mouth. Company 
A moved east along the coastal road back 
toward the perimeter while the reinforced 
platoons of Company L traveled over the 
difficult terrain inland from the beach. 
Company A met brief Japanese resistance 
near the mouth of the river, but neither 
force found a trace of the Goettge patrol." 

This action was followed a week later 
by a planned double envelopment against 
the village of Matanikau. Companies B 
and L of the 5th Marines would carry out 

"" Goettge's position as Division G-2 was filled 
on 14 August by LtCol E. J. Buckley, formerly of 
the 11th Marines. FinalRept, Phase III, Annex 
F, 5. 



MARINE ENGINEERS erected this bridge across the Tenaru River to speed the flow of troops and sup- 
plies from the beaches to the perimeter. ( USMC 50468-A) 

SOLOMONS NATIVES, recruited by Captain W. F. Martin Clemens, BSIDF, guide a patrol up the 
winding course of the Tenaru River in search of Japanese. ( USMC 53325 J 



this attack while Company I of the same 
regiment made an amphibious raid farther 
west, at Kokumbona, where it was hoped 
that any Japanese retreating from Matani- 
kau could be cut off. On 18 August 
Company L moved inland, crossed the 
Matanikau some 1,000 yards from the 
coast, and prepared to attack north into 
the village the next day. Company B, to 
attack west, moved along the coastal road 
to the east bank of the river. 

Next day, after preparation fire was laid 
down by the 2d, 3d, and 5th Battalions of 
the 11th Marines, Company L launched its 
attack. Shortly after jumping off, scouts 
discovered a line of emplacements along a 
ridge some 1,000 yards to the left flank of 
the company front. The platoon on this 
flank engaged the small enemy force in 
these emplacements while the remainder of 
the company moved on toward the village. 
In this action off the left flank. Sergeant 
John H. Branic, the acting platoon leader, 
was killed. The company executive offi- 
cer, Lieutenant George H. Mead, Jr. next 
took command. When he was killed a 
short time later Marine Gunner Edward 
S. Rust, a liaison officer from the 5th Ma- 
rines headquarters took command. This 
platoon continued to cover the advance of 
the remainder of the company. 

Company B, thwarted in its attempt to 
cross the river because of intense Japanese 
fire from the west bank, could only sup- 
port the attack of its sister company by 
fire. Company L managed to reach the 
outskirts of the native village at about 
1400, however, and one platoon entered the 
settlement. This platoon lost contact with 
the remainder of the company, and when 
the other platoons attempted to enter the 
village they were met by a strong Japanese 
counterattack which caused the separated 
platoon to withdraw to company lines. 
The Marines were nearly enveloped and 

succeeded in repulsing the attack only 
after close-range fighting. Defending in 
depth, the Japanese drew up on a line 
which extended from the river some 200 
yards west through the village. While 
the Marines maneuvered for an attack, the 
Japanese fire became more sporadic, and 
an assault at about 1600 revealed that the 
enemy pocket had dispersed. 

Meanwhile the amphibious raid of Com- 
pany I also aroused opposition. Two 
enemy destroyers and a cruiser lobbed 
shells at the landing craft while they 
swung from the Marine perimeter to Ko- 
kumbona, and the raiding party escaped 
this threat only to be met at the beach 
by Japanese machine-gun fire. The land- 
ing succeeded, however, and the enemy re- 
sistance began to melt. By the time a 
Marine attack swept through the village, 
the defenders had retired into the jungle 
to avoid a conclusive engagement. The 
three companies killed 65 Japanese while 
suffering the loss of four Marines killed 
and 11 wounded. 

Although these actions served only to 
locate the general area into which the 
original Japanese garrison of Guadal- 
canal had withdrawn in the face of the 
Marine landing, another patrol on 19 Au- 
gust indicated the pattern of things to 
come on the island. 

The patrol and reconnaissance area as- 
signed to the 1st Marines lay east and 
southeast of the perimeter where the plains 
of the Lunga fan into a grassy tableland 
which is nearly eight miles wide near the 
coastal village of Tetere. Some thought 
had been given to the construction of an 
airfield there, and on 12 August a survey 
party went out with a platoon-sized secur- 
ity force under Second Lieutenant John 
J. Jachym. Passing through a native vil- 
lage on 13 August, this group encountered 
Father Arthur C. Duhamel, a young Cath- 





olic priest from Methuen, Massachusetts," 
who related native rumors of an enemy 
force along the coast to the east. 

Two days later a partial verification of 
the priest's information was made by Cap- 
tain (of the British Solomon Islands De- 
fense Force) W. F. M. (Martin) Clemens, 
the district officer who had withdrawn into 
the hills to become a coastwatcher when the 
Japanese entered his island. On 14 
August Clemens left his watching station 
near Aola Bay with his 60 native scouts " 
and entered the Marine perimeter. 
Clemens and his scouts reported seeing 
signs of a new Japanese force. And on 
the heels of Clemens' reports came word 
from Admiral Turner that naval intelli- 
gence indicated a Japanese attack in force. 

To investigate, Captain Charles H. 
Brush, Jr. took a part of his Company A 
of the 1st Marines and at 0700 on 19 
August began a patrol eastward along the 
coastal track toward Koli Point and 
Tetere. At about noon near Koli Point 
the patrol spotted a group of four Jap- 
anese officers and 30 men moving, with no 
security to front or flanks, between the 
road and the beach. Captain Brush struck 
frontally with a part of his unit while 
Lieutenant Jachym led an envelopment 
around the enemy left flank. In 55 min- 
utes of fighting, 31 of the Japanese were 
killed. The remaining three escaped into 

" Father Duhamel, as well as Father Henry 
Oude-Engberink and Sister Sylvia of France, and 
Sister Odilla of Italy, were later tortured and 
killed by the Japanese. 

" Including Vouza, a retired sergeant major 
of police and native of the Tetere area who 
previously had volunteered additional service to 
the Crown and Capt Clemens. A veteran of 25 
years in the native constabulary, the "reacti- 
vated" Vouza provided valuable assistance to the 
coastwatchers and to the Marines. 

the jungle. Three Marines were killed 
and three wounded. 

It was clear that these troops were not 
wandering laborers or even members of 
the original garrison. Helmets of the 
dead soldiers bore the Japanese Army star 
rather than the anchor and chrysanthe- 
mum device of the special landing force. 
A code for ship-to-shore communications 
to be used for a landing operation also was 
found among the effects, and the appear- 
ance of the uniforms indicated that the 
troops were recent arrivals to Guadalcanal. 
There appeared little doubt that the Japa- 
nese were preparing an attempt to re- 
capture their lost airfield. And Brush 
found they had already completed some 
excellent advance work : 

With a complete lack of knowledge of Japanese 
on my part, the maps the Japanese had of our 
positions were so clear as to startle me. They 
showed our weak spots all too clearly." 

While these patrols searched for the 
enemy on Guadalcanal, another force of 
approximately 200 Marines moved into 
enemy waters farther north and raided a 
Japanese atoll in the Gilbert Islands. 
Companies A and B of Lieutenant Colonel 
Evans F. Carlson's 2d Raider Battalion 
went from Oahu to Makin atoll on board 
submarines Argonaut and Nautilvs and 
landed on the hostile beach early on 17 Au- 
gust. The raid was planned to destroy 
enemy installations, gather intelligence 
data, test raiding tactics, boost homef ront 
morale, and possibly to divert some Japan- 
ese attention from Guadalcanal. It was 
partially successful on all of these counts, 
but its greatest asset was to home-front 
morale. At a cost to themselves of 30 men 
lost, the raiders wiped out the Japanese 
garrison of about 85 men, destroyed radio 

' Maj C. H. Brush, Jr. Itr to CMC, 15Jan49. 



stations, fuel, and other supplies and in- 
stallations, and went back on board their 
submarines on 18 August for the return 
to Pearl Harbor. This raid attracted 
much attention in the stateside press but 
its military significance was negligible. 
Guadalcanal still held the center of the 
stage in the Pacific and attention quickly 
turned back to that theater." (See Map 


The picture that began to take shape as 
these bits of intelligence fitted together 
provided an early warning of Japanese 
plans that already were well underway. 
On 13 August, Tokyo ordered Lieutenant 
General Haruyoshi Hyakutake's Seven- 
teenth Army at Rabaul to take over the 
ground action on Guadalcanal and salvage 
the situation. The naval side of this rein- 
forcement effort would be conducted by 
Rear Admiral Raizo Tanaka, a wily Im- 
perial sea dog who was a veteran of early 
landings in the Philippines and Indonesia 
and of the battles of Coral Sea and Mid- 
way. With no clear picture of his oppo- 
nent's strength, Hyakutake decided to re- 
take the Lunga airfield immediately with 
a force of about 6,000 men. On the eve- 
ning of the 15th of August, while Tanaka's 
ships of the reinforcement force were load- 

"CTF 7.15 Rept, 24Aug42; 2dRdrBn Rept of 
Ops on Makinis, 19Aug42 ; WDC Japanese Docu- 
ments No. 161,013, 161,110, and NA 12053, "Rec- 
ords of Various Base Forces" and "Base Force 
Guard Units and Defense Unit Records," 17-22- 
Aug42 (located at NHD). 

" Actually the Ilu. But as previously ex- 
plained. Marines of the division identified these 
rivers incorrectly throughout the campaign and 
the action to be described has thus become known 
historically and to the participants as the Battle 
of the Tenaru. 

ing supplies at Truk, the admiral got 
orders to hurry down to Rabaul and take 
900 officers and men to Guadalcanal at 
once. Hyakutake had decided that the 
attack would begin with a part of the 7th 
Division's 28th Infantry Regiment and the 
Yokosuka Special Naval Landing Force. 
These units would be followed by the 36th 

Admiral Tanaka thought he was being 
pressed a little too hard, considering that 
the Eighth Fleet under which he operated 
had just been formed at Rabaul on 14 July, 
and that the admiral himself had hardly 
been given time to catch his breath after 
hurrying away from Yokosuka for his new 
job on 11 August. The admiral reported 
later : 

With no regard for my opinion . . . this 
order called for the most difficult operation in 
war — landing in the face of the enemy — to be 
carried out by mixed units which had no oppor- 
tunity for rehearsal or even preliminary study. 
... In military strategy expedience sometimes 
takes precedence over prudence, but this order 
was utterly unreasonable. 

I could see that there must be great confusion 
in the headquarters of Eighth Fleet. Yet the 
operation was ordained and underway, and so 
there was no time to argue about it." 

Backbone of the initial effort would con- 
sist of the reinforced 2d Battalion., 28th 
Infantry., a 2,000-man force of infantry, 
artillery, and engineers under the com- 
mand of Colonel Kiyono Ichiki. This 
force had been en route to Midway when 
the defeat of the Japanese carriers caused 
a change to Guam." Later the Ichiki 
Force was en route back to the home is- 
lands when the Marine landing in the 
Solomons brought another change of 

" Tanaka Article, I, 690. Excerpts from this 
account are quoted in this volume with the per- 
mission of the U. S. Naval Institute. 

'' See Part V of this volume. 





Japanese plans. The unit was diverted to 
Truk where it landed on 12 August and 
was attached to the 36th Brigade which 
then garrisoned the Palau Islands. The 
brigade's commander was Major General 
Kiyotake Kawaguchi. The 900 or 1,000 
men which Admiral Tanaka loaded for his 
first reinforcement run to Guadalcanal 
were from this Ichiki unit. 

The reinforcement ships landed Colonel 
Ichiki and this forward echelon at Taivu 
Point on Guadalcanal during the night of 
18 August. While this force landed at 
this point some 22 air miles east of the 
Lunga, some 500 men of the Yokosuka 
Fifth Special Naval Landing Force ar- 
rived at Kokumbona. This was the first 
of many runs of the Tokyo Express, as 
Marines called the Japanese destroyers 
and cruisers which shuttled supplies and 
reinforcements up and down The Slot in 
high-speed night runs. Brush's patrol 
had encountered part of Ichiki's forward 
echelon, and the Japanese commander, 
shaken by the fact that he had been dis- 
covered, decided to attack at once. 

At that time the Marines had five in- 
fantry battalions available for defense of 
Lunga Point. Four battalions were com- 
mitted to beach defense, one was withheld 
in division reserve. On 15 August work 
had begun on a new extension of the right 
flank by refusing it inland along the west 
bank of the Ilu River (then called the 
Tenaru) for a distance of 3,200 yards. 
This plan involved road and bridge con- 
struction as well as extensive clearing be- 
fore field fortifications could be built. As 
of 18 August little progress had been 
made. (See Map 17) 

In the face of the threats pointed out by 
intelligence sources, the division consid- 
ered two courses of action : first, to send 
the division reserve across the Ilu to locate 

and destroy the enemy, or, second, to con- 
tinue work on defensive positions while 
limiting actions to the east to strong pa- 
trols and outposts. The first course. Gen- 
eral Vandegrift realized, involved accept- 
ing the premise that the main Japanese 
force had landed to the east and that it 
could be dealt with by one Marine bat- 
talion. But if Brush's patrol had en- 
countered only a small part of the new 
enemy unit while the bulk of the force 
stood poised to strike from another direc- 
tion, or from the sea, absence of the reserve 
battalion would become a serious man- 
power shortage in the perimeter. The 
intelligence Vandegrift had gleaned from 
all sources was good, but there wasn't 
enough of it. So the division sat tight to 
await developments. Work continued on 
field fortifications, native scouts worked 
far to the east, and Marines maintained a 
strong watch on the perimeter each night. 
The Marines did not have long to wait. 
Colonel Ichiki had wasted no time pre- 
paring his attack, and during the night 
of 20-21 August Marine listening posts on 
the east bank of the Ilu detected enemy 
troops moving through the jungle to their 
front. A light rattle of rifle fire was ex- 
changed, both sides sent up flares, and the 
Marines withdrew across the river mouth 
to the lines of their battalion. Lieutenant 
Colonel Edwin A. Pollock's 2d Battalion, 
1st Marines. They reported that a strong 
enemy force appeared to be building up 
across the river .^* 

" At about the same time the native scout 
Vouza entered the command post of 2/1 to warn 
LtCol I'ollock about the Japanese buildup. Badly 
wounded, Vouza had been captured by the Ichiki 
Force, knifed about the face, throat, and chest 
when he wouldn't talk, and then left for dead by 
the .Japanese. This report to Pollock was one of 
the many services for which Vouza later was cited 
by the Americans and British. 



o >" 


S T- o- 




£ • 


U ^ 


— » 



"° £ 

• w 


nt wh 



on po 


u. 3 





•1 i 

r ^ 


< -5 o 

" ° ^ 








83My nyVNSI 



By this time Ichiki had assembled his 
force on the brush-covered point of land 
on the east bank of the river, and all was 
quiet until 0310 on 21 August when a 
column of some 200 Japanese rushed the 
exposed sandspit at the river mouth. Most 
of them were stopped by Marine small- 
arms fire and by a canister-firing 37mm 
antitank gun of the 1st Special Weapons 
Battalion. But the Marine position was 
not wired in, and the weight of the rush- 
ing attack got a few enemy soldiers into 
Pollock's lines where they captured some 
of his emplacements. The remainder of 
the line held, however, and fire from these 
secure positions kept the penetration in 
check until the battalion reserve could get 
up to the fight. This reserve, Company G, 
launched a counterattack that wiped out 
the Japanese or drove them back across 
the river. 

Ichiki was ready with another blow. 
Although his force on the east bank had 
not directly supported this first attack, it 
now opened up with a barrage of mortar 
and 70mm fire, and this was followed by 
another assault. A second enemy company 
had circled the river's mouth by wading 
beyond the breakers, and when the fire 
lifted it charged splashing through the 
surf against the 2d Battalion's beach posi- 
tions a little west of the river mouth. 

The Marines opened up with everything 
they had. Machine-gun fire sliced along 
the beach as the enemy sloshed ashore, 
canister from the 37mm ripped gaping 
holes in the attack, and 75mm pack howit- 
zers of the 3d Battalion, 11th Marines 
chewed into the enemy. Again the attack 
broke up, and daylight revealed a sandy 
battlefield littered with the bodies of the 
Japanese troops who had launched Gua- 
dalcanal's first important ground action. 

Although outnumbered at the actual 
point of contact, Pollock assessed the situ- 
ation at daybreak and reported that he 
could hold. His battalion had fire superi- 
ority because of the excellent artillery 
support and because the course of the river 
gave part of his line enfilade fire against 
the enemy concentration in the point of 
ground funneling into the sandspit. In 
view of this. General Vandegrift ordered 
Pollock to hold at the river mouth while 
the division reserve, the 1st Battalion, 1st 
Marines enveloped Ichiki. While this bat- 
talion prepared for its attack. Company 
C of the 1st Engineer Battalion went for- 
ward to Pollock's command to help bolster 
defensive positions. During the morning 
the engineers built antitank obstacles, laid 
a mine field across the sandspit, and helped 
the 2/1 Marines string tactical wire and 
improve field fortifications. They were 
under intermittent rifle fire during most 
of this work. 

Meanwhile, Lieutenant Colonel Lenard 
B. Cresswell's division reserve battalion 
had reverted to parent control and re- 
ported to Colonel Cates to receive the 
attack plan for the envelopment. Before 
0700, Cresswell crossed the Ilu upstream, 
posted elements of his Company D (weap- 
ons) to cover a possible Japanese escape 
route to the south, and then turned north 
toward the Ichiki Force. By 0900 his com- 
panies crossed their lines of departure in 
the attack against the Japanese left and 

Company C on the right along the coast 
met one platoon of the enemy near the 
village at the mouth of the Block Four 
River, and the Marines moved to encircle 
this force and isolate it from the remainder 
of Ichiki's unit farther west. The other 
companies moved north with little opposi- 



tion, A on the right and B on the left. As 
the advance continued, the enemy was 
forced into the point of land on the Ilu's 
east bank. By 1400 the enemy was con- 
fined completely by the river, the beach, 
and the envelopment from the left and 
rear. Some of the Japanese made un- 
successful attempts to escape through the 
surf and along the beach; another group 
burst out temporarily to the east but ran 
head-on into Company C moving up from 
its battle at the mouth of the Block Four. 

The fight continued, with Cresswell 
tightening his encirclement, and more of 
the Japanese attempted to strike through 
to the east. These breakout attempts gave 
the new Guadalcanal fliers, on the island 
less than 24 hours, a chance to fire their 
first shots in anger, and the F4F pilots 
from VMF-223 gave Cresswell's Marines 
a hand with strafing attacks that destroyed 
the Japanese or turned them back into the 
infantry trap. 

To conclude the action by nightfall, 
Vandegrift ordered a tank attack across 
the sandspit and into what now had be- 
come the rear of the Ichiki Force. The 
platoon of light tanks struck at 1500, firing 
at the enemy with canister and machine 
guns. Two tanks were disabled, one by 
an antitank mine, but the crews were 
rescued by the close supporting action of 
other tanks and the attack rolled on into 
the Japanese positions. It was over by 
1700. Nearly 800 Japanese had been killed 
and 15 were taken prisoner while only a 
few escaped into the jungle. Disgraced 
by the debacle. Colonel Ichiki committed 

The action cost the Marines 34 dead and 
75 wounded. A policing of the Japanese 
battlefield gleaned the division ten heavy 
and 20 light machine guns, 20 grenade 

throwers, 700 rifles, 20 pistols, an unde- 
termined number of sabers and grenades, 
three 70mm guns, large quantities of ex- 
plosive charges, and 12 flame throwers. 
The flame throwers were not used in the 

Admiral Tanaka later had this to say 
about the disaster: 

I knew Colonel Ichiki from the Midway opera- 
tion and was well aware of his magnificent lead- 
ership and indomitable fighting spirit. But this 
episode made It abundantly clear that Infantry- 
men armed with rifles and bayonets had no chance 
against an enemy equipped with modern heavy 
arms. This tragedy should have taught us the 
hopelessness of 'bamboo-spear' tactics. " 


While Colonel Ichiki prepared for his 
ill-fated attack. Rear Admiral Tanaka and 
Vice Admiral Gunichi Mikaw-a, the Eighth 
Fleet commander, worked to get the colo- 
nel's second echelon ashore for what they 
hoped would be an orderly, well-coordi- 
nated effort against the Marines. These 
troops were on board the Kinryu Maru, and 
four destroyer transports, and they were 
escorted by the seaplane carrier Chitose 
with her 22 floatplanes and by Tanaka's 
Destroyer Squadron 2, which Tanaka led 
in light cruiser Jintsu. A larger naval 
force operated farther to the east outside 
the Solomons chain. In all, the Japanese 
task forces included three aircraft carriers, 
eight battleships, four heavy cruisers, two 
light cruisers, and 22 destroyers in addi- 
tion to the five transport vessels. 

At this time Admiral Fletcher's force 
of two carriers, one battleship, four 
cruisers, and ten destroyers operated to the 
southeast of the Lower Solomons conduct- 

Tanaka Article, I, 691. 



ing routine searches to the northwest. 
Fletcher believed the area to be tempo- 
rarily safe from Japanese naval trespass, 
and he had sent the carrier Wasp off to 
refuel. This left him only the Enterprise 
and the Saratoga for his air support. 

On 23 August, two days after the Battle 
of the Tenaru, American patrol planes first 
sighted the Japanese transports and the 
Tanaka escort some 350 miles north of 
Guadalcanal. Marine planes from Hen- 
derson Field attempted to attack the troop 
carriers, but a heavy overcast forced them 
back to Lunga. The fliers had a better day 
on the 24th, however. At 1420 the F4F 
pilots intercepted 15 Japanese bombers 
being escorted toward Guadalcanal by 12 
fighters from the carrier Ryujo. Marines 
broke this raid up before it got close. 
They downed six of the Zeros and ten 
bombers in what was VMF-223's first big 
success of the war. Captain Marion Carl 
splashed two bombers and a Zero, and two 
planes each were downed by Lieutenants 
Zennith A. Pond and Kenneth D. Frazier, 
and Marine Gunner Henry B. Hamilton : 

This was a good day's work by the fighter 
pilots of VMF-223. It is necessary to remember 
that the Japanese Zero at this stage of the war 
was regarded with some of the a^e in which the 
atomic bomb came to be held later. . . . The 
Cactus [Guadalcanal] fighters made a great con- 
tribution to the war by exploding the theory that 
the Zero was invincible ; the Marines started the 
explosion on 24 August."" 

Three Marine pilots did not return from 
the action, and a fourth was shot down but 
manttged to save himself by getting ashore 
at Tulagi. In plane strength, however, 
the Cactus Air Force (as the Guadalcanal 
fliers called their composite outfit) gained. 
This was the day, as mentioned earlier in 
this chapter, that the 11 SBD's came in 

' ilarinc Air History, 81. 

from the damaged Enterprise. At the 
time the ship was struck. Lieutenant 
Turner Caldwell, USN, was up with his 
"Flight 300," and, low on gas, he led his 
fliers to Guadalcanal where they more than 
paid for their keep until 27 September. 

Meanwhile Admiral Fletcher's carrier 
planes located the enemy task force in the 
Eastern Solomons at about the same time 
Japanese planes spotted Fletcher. Like 
the Battle of Midway, the resulting action 
was an air-surface and air-air contest. 
Surface vessels neither sighted nor fired at 
each other. 

The Ryujo, whose Zeros had fared so 
poorly with John Smith's F4F pilots, took 
repeated hits that finally put her out of 
control and left her hopelessly aflame. 
One enemy cruiser and a destroyer were 
sunk; a second cruiser was damaged; the 
Ohitose sustained severe wounds but man- 
aged to limp away ; and 90 Japanese planes 
were shot down. On the American side, 
20 planes were lost and the damaged 
Enterprise lurched away to seek repair. 

This action turned back the larger Jap- 
anese attack force, and Fletcher likewise 
withdrew. He expected to return next day 
and resume the attack, but by then the 
Japanese had moved out of range. The 
escorted transports with reinforcements 
for the late Colonel Ichiki continued to 
close the range, however, and early on 25 
August SBD's from VMSB-232 and the 
Enterprise Flight 300 went up to find 
them. The Battle of the Eastern Solo- 
mons had postponed Tanaka's delivery of 
these reinforcements, but after that carrier 
battle was over the admiral headed his 
ships south again late on 24 August. 

At 0600 on 25 August, Tanaka's force 
was some 150 miles north of Guadalcanal, 
and thei-e the SBD's from Henderson 



Field found him. The Jintsu shook under 
an exploding bomb that Lieutenant Law- 
rence Baldinus dropped just forward of 
her bridge, and Ensign Christian Fink of 
the Enterprise scored a hit on the trans- 
port Kinryu Maru amidships. Admiral 
Tanaka was knocked unconscious by the 
explosion on his flagship, and a number of 
crewmen were killed or injured. The ship 
did not list under the bow damage, how- 
ever, and she still was seaworthy. When 
Tanaka recovered he transferred his flag 
to the destroyer Kagero and sent the 
Jintsu to Truk alone.^^ 

Flames broke out on the Kinryu Maru 
which carried approximately 1,000 troops 
of the Yokosuka 6th SNLF, and the de- 
stroyer Muzuki went alongside to rescue 
survivors. At just that moment this ship 
became "one of the first Japanese warships 
to be hit by a B-17 since the war began" ^^ 
when these big planes from the 11th Bom- 
bardment Group at Espiritu Santo arrived 
to lend a hand to the Cactus fliers. The 
Muzuki sank at once. Another ship then 
moved in to rescue the survivors from this 
destroyer while two destroyer transports 

" Tunaka Articlei 1, 693. 

"Marine Air History 81. See also Tanaka 
Article, I, 694 ; Struggle for Guadalcanal, 105. 

went to the rescue of the men from the 
Kinryu Maru. These men were picked up 
just as the Maru also went to the bottom. 
Meanwhile another pass at the ships had 
resulted in light damage to the destroyer 
UzuJci, and Admiral Tanaka turned back 
for Rabaul. Many of the SNLF men had 
been lost, and his force was badly shaken 
and disordered : 

My worst fear for this operation had come to 
be realized. Without the main combat unit, the 
Yoltosuka 5th Special Naval Landing Force, it 
was clear that the remaining auxiliary unit of 
about 300 men would be of no use even if it did 
reach Guadalcanal without further mishap." 

Thus had the 1st Marine Division 
gained some valuable time to prepare for 
the next Japanese attempt to dislodge its 
Lunga defense. With air support on Hen- 
derson Field and with a tenuous supply 
route established to the New Hebrides, the 
division's grip on Guadalcanal was much 
improved at month's end. But it still was 
a long way from being completely secure, 
especially now that Ichiki's act of hara- 
kiri had pointed up for the Japanese the 
impropriety of trying to dislodge the land- 
ing force with only 900 or 1,000 men. 

' Tanaka Article, I, 694. 

The Battle of the Ridge 


General Vandegrift and his staff were 
aware that the defeat of the Ichiki Force 
left the division's position on the island 
only temporarily improved. Obviously 
the Japanese could be expected to mount 
larger and better planned attacks against 
the small Marine perimeter ; air and naval 
activity at Guadalcanal indicated no 
waning enemy interest in the South Solo- 
mons area. A noon-hour visit from Ra- 
baul bombers was an almost daily occur- 
rence, and enemy warships and submarines 
entered Sealark Channel nearly every 
night to shell Henderson Field. 

Although the Battle of the Eastern 
Solomons gave Allied shipping from Es- 
piritu Santo an opportunity to increase 
the flow of supplies to the beleaguered Ma- 
rines, the Lunga defenders still operated 
on a hand-to-mouth basis. 

The Cactus Air Force performed beyond 
all proportion to its facilities and equip- 
ment, and the 3d Defense Battalion finally 
was able late in August to bring in the 5- 
inch guns of its two seacoast batteries; 
but there were not enough Marines on the 
island to enlarge the perimeter for an ade- 
quate defense. General Vandegrift be- 
lieved that positions along 45 miles of 
Guadalcanal's north coast would have to be 
held before the Japanese could be re- 
strained from landing and attacking Hen- 
derson Field and before air defenses would 
have sufficient room for deployment. The 
general did not have that kind of man- 

Since Major John Smith and Lieuten- 
ant Colonel Mangrum arrived with their 
F4F's and SBD's on 20 August, the airfield 
had taken on a more proficient and perma- 
nent look. By the end of August a daily 
routine of scheduled patrol flights had 
been initiated. Four-plane fighter patrols 
flew from 0545 to 0830 each morning and 
from 1400 to 1830 each afternoon, and 
mixed fighter-bomber squadrons fre- 
quently made night searches for enemy 
shipping to the northwest. Cactus avia- 
tors flew cover for the Allied shipping to 
the island, and went up on intercept during 
the Japanese raids. 

The U. S. fighters did well against the 
enemy bombers, but their only chance 
against the highly maneuverable Zero was 
to pair up in mutual support. In this way 
they could protect themselves when the 
Zeros came down to drive them away from 
the bombers. They found that the Grum- 
man did have certain advantages over the 
Zero, however. It had great fire power, 
and it could stay in the air with more holes 
in it than the more flimsy Japanese fighter 
could endure. During the first ten days 
of Cactus operations, U. S. fliers shot down 
56 Japanese planes at a cost of 11 of their 
own craft. 

Marine engineers rigged a system of 
lights from captured Japanese equipment 
to outline the field for emergency night 
landings, and, when dump trucks and 
pneumatic tampers came in later, workers 
could fill a 500-kilogram-bomb crater in 30 




minutes. Dump trucks were kept loaded 
with gravel and sand, and "flying squads" 
of engineers rushed out to repair any dam- 
age immediately after the departure of 
Japanese bombers. 

But not even counting enemy action, 
Henderson personnel still had plenty of 
problems. An early method of fueling 
employed drums strung up in the rafters 
of partially built Japanese hangars, and 
even when gasoline trucks arrived later 
the fuel had to be hand pumped from 
drums to the trucks. There was no steel 
matting, and the field was completely at 
the mercy of the whimsical tropical 
weather : 

Henderson Field was a bowl of black dust 
which fouled airplane engines or it was a quag- 
mire of black mud which made the take-off 
resemble nothing more than a fly trying to rise 
from a runway of molasses. ' 

When engineers and Seabees had no 
bomb craters to patch, they still had to fix 
up the field in the wake of the early SBD's 
which had hard-rubber tail wheels de- 
signed for landing on the sturdy decks of 
carriers. On the Henderson earth these 
wheels ". . . chewed up the runway like 
a plowshare." ^ The sorry condition of 
the field added serious operational losses 
to the troubles of the small Cactus force 
which was nearly always outnumbered in 
the air. Occasionally a plane was gripped 
so persistently by the mud that it failed 
to take off and crashed at the end of the 
runway ; ruts and the beginnings of pot- 
holes were hazards on dry days, and on 
one foggy wet afternoon in early Septem- 
ber a landing F4F crashed into a bulldozer. 

But in spite of everything the installa- 
tion grew and slowly improved, and this 

was a period when American fighting men 
were thankful for small favors. On 20 
August the transport William Ward Bvr- 
rows came up from the New Hebrides with 
the forward echelon of MAG-23. All the 
men and some of the gear were put ashore, 
but then the ship scurried across Sealark 
Channel for Tulagi when the word came in 
that a Japanese cruiser force was expected 
that night. Near Tulagi the transport 
went aground and much of the equipment 
still on board had to be jettisoned to float 
her free. 

Next day Colonel William J. Wallace, 
group commander, came up to Henderson 
with more planes : 19 F4F's of Major 
Kobert E. Galer's VMF-224, and 12 
SBD-3's of VMSB-231 commanded by 
Major Leo R. Smith. That brought the 
Cactus strength to 86 pilots and 64 planes, 
10 of them Navy and three Army. 

On 1 September the ground crews got 
more help. Five officers and 387 men of 
the 6th Naval Construction Battalion 
(Seabees) landed with two bulldozers. 
They would ". . . help make an airfield 
out of Henderson and . . . clear a short 
grassy strip a mile to the east called 
Fighter 1." ^ But next day came one of 
the infamous Henderson disasters that 
always loomed as a threat to much of the 
backbreaking effort that had gone before. 

With the frequent raids, fire was always 
a dangerous possibility, and a field fire 
brigade had been organized around two 
Japanese trucks which had been repaired 
by the 1st Marines. They got their bap- 
tism on 2 September when a bomb from a 
heavy Japanese raid hit an armed SBD 
parked at the edge of a coconut grove 
where ammunition was stored. The bomb 
could not be removed from the burning 

^ Marine Air History, 82. 
= /6td., 83. 

' IMd., 84. 




90MM ANTIAIRCRAFT GUNS of the 3d Defense Battalion point skyward at Henderson Field on alert 
against the attacks of Japanese bombers. (USMC 61608) 

1 05 MM HOWITZER of the 11th Marines nestles beneath the slope of a protecting ridge on Guadalcanal 
ready to go into action against the Japanese. (USMC 51832 ) 



SBD, and when it exploded it spewed 
flaming gasoline in all directions. One 
90mm shell dump was ignited, and the fire 
brigade could not do its best work with 
all the explosions that resulted. Several 
of the fire-fighters were injured, and the 
trucks seemed to be making little headway 
since they had to take turns dashing off to 
the Lunga Kiver, the closest supply of 
water. If the fire expanded much more 
it would set off a chaiii reaction and all 
the ammunition in the area would be lost. 

Had not the situation been so grim, some 
old hands might have been reminded of the 
Chinese fire drill of ancient Marine legend. 
The blaze was eventually brought under 
control, however, and the loss was serious 
but not critical. After this, large water 
tanks from coconut plantations were 
spotted around the ammunition dumps; 
but this fire proved to be the most serious 
of the campaign. Subsequent losses 
occurred in division dumps as a result of 
naval shelling at night. These losses were 
negligible since the ammunition by that 
time had been buried. 

This bombing raid had arrived at 1135, 
and while the fire department below 
worked to save the ammunition dumps. 
Cactus fliers were up among the bombers. 
They shot down three of the twin-engined 
craft and four Zeroes without a single loss 
of their own. 

On 3 September the command echelon 
of the 1st Marine Aircraft Wing arrived. 
This group included Brigadier General 
Roy S. Geiger, commanding general of the 
wing; his chief of staff. Colonel Louis E. 
Woods; and Lieutenant Colonel John C. 
Munn, wing intelligence officer. Using the 
MAG-23 staff as his wing staff, Geiger 
established his command post near that 
of General Vandegrift. Liaison in the 
form of daily conferences between the two 
generals was established, and Kenneth H. 
Weir, now a lieutenant colonel, at last had 

a well-organized air headquarters with 
which to deal as division air officer. * 

Fueling and arming of the planes con- 
tinued in a make-shift manner for some 
time, and as late as November bombs had 
to be manhandled. Eadio communications 
likewise posed problems. Army and Navy 
receiving channels did not mesh, and the 
Army planes of the Cactus Force could not 
receive Navy traffic. Operations resolved 
this by employing the radio from a 
grounded Army P-400 alongside the Navy 
set and thereafter making simultaneous 
broadcasts over twin microphones. This 
was a big help, but communications still 
were far from satisfactory. Beyond 20 
miles the planes could not depend on re- 
ceiving the field, but the field could nor- 
mally read the planes' messages from as 
far as 100 miles. 

Since the fight against Ichiki, there had 
been little opportunity for close air support 
of ground troops, but Marines continued 
to plan for this sort of air-ground team- 
work. Communications was the big prob- 
lem here, too. At that stage of operations 
only visual signals were used, consisting 
mainly of colored panels which the ground 
troops had, but they left much to be de- 
sired. Planes now flew higher and faster 
than they had in the banana wars and 
maneuvers, and this made it more difficult 
for pilots to read the panel messages, even 
if they could catch a glimpse of the colored 
markings. And more often than not in 
Guadalcanal's thick jungle and tall grass, 
they could not even see them. Guadal- 
canal Marines had heard about colored 
smoke grenades which were being tested 
back in the States, and they thought these 

' As mentioned in the previous chapter, Weir's 
first look at the field had come on 8 August when 
he and the division engineer oflBcer had esti- 
mated hovv' much work they would have to add to 
the early Japanese efforts to make the strip 



might be helpful foi' air-ground signals. 
But what they really had their eyes out for 
were some radio sets. That seemed to be 
the only promising solution for air-ground 
coordination. Radios initially available 
to the division would not serve the pur- 
pose, and it would not be until October 
that Vandegrift could detail an officer 
and suitable radio equipment and person- 
nel to train as "air forward observers" 
from each infantry regiment and thereby 
pioneer in what later became an important 
phase of Marine combat operations. 

While Geiger built up his air arm, 
Vandegrift likewise added strength to the 
Lunga perimeter. With Tulagi quiet, he 
brought some of General Rupertus' troops 
across the, channel to Guadalcanal. The 
2d Battalion, 5th Marines made the move 
on 21 August, and the 1st Raider Battalion 
and the 1st Parachute Battalion crossed 
to the Guadalcanal side on 31 August. In 
early September, when a detachment of the 
5th Defense Battalion came ashore at 
Tulagi, a 90mm battery of the 3d Defense 
Battalion joined its parent organization in 
the Henderson Field area. 

From all indications these additional 
troops would be needed. Aerial observa- 
tion and native scouts piled up reports of 
Japanese landings on both sides of the 
perimeter, and staff officers estimated a 
build-up of some 200 or 300 well-equipped 
enemy troops near the village of Tasim- 
boko some 18 miles east of Lunga Point. 
Native scouts placed the enemy strength 
much higher, but Marines suspected such 
counts to be exaggerated. 

Patrolling continued in all sectors, and 
on 27 August the 1st Battalion, 5th Ma- 
rines under Lieutenant Colonel William E. 
Maxwell met a strong body of troops near 
the village of Kokumbona, west of the 
perimeter. The battalion had made an 

amphibious landing without incident at 
about 0730, but later ran into the Japanese 
force dug into positions throughout a 
narrow coastal gorge. Maxwell was be- 
yond artillery range of the perimeter, and 
although the 2d and 5th Battalions of the 
11th Marines fired diversionary missions 
east of him in Matanikau village, the Jap- 
anese facing the infantry Marines seemed 
inclined for once to make a strong stand 
rather than to slink off into the brush as 
they had frequently done in other such 

Faulty communications and other diffi- 
culties bogged the Marine attack, and 
Lieutenant Colonel Maxwell withdrew his 
force to comply with a portion of his patrol 
order which required him to return to the 
perimeter by nightfall. But the regi- 
mental commander, Colonel Hunt, ordered 
the battalion back into the fight, relieved 
Maxwell of command, and soon thereafter 
arrived on the scene himself. Major Mil- 
ton V. O'Connell succeeded to command 
of the battalion, but the attack was not 
resumed until the predawn hours of the 
following morning. A few Japanese were 
killed, but most of them had withdrawn. 
The Marines retired to Matanikau vil- 
lage and later returned by water to the 

On 2 September two companies of the 
raider battalion patrolled Savo Island 
but found no enemy. Following this the 
raiders and parachutists, consolidated into 
a provisional battalion, moved into defen- 
sive positions on the south rim of the per- 
imeter, inland from the airfield. While 
they dug in. Colonel Edson and his staff 
made plans for an amphibious raid to the 
east where the enemy build-up was re- 
ported around the Tasimboko area. 

The landing was made just east of 
Tasimboko before dawn on 8 September, 



and the raiders ^ advanced west into the 
rear of the reported Japanese positions. 
At about 0630 planes of MAG-23 bombed 
and strafed the suspected strong point, and 
two destroyer transports, Manley and Mc- 
Kean, opened up on the area. At 0830 
Edson made contact against light resist- 
ance, and his advance overran two artil- 
lery pieces. He still could not determine 
the strength of the enemy, but the force 
appeared to be withdrawing toward the 
village, and he requested that supporting 
dive bombers remain on station in the 
event that the enemy pocket could be local- 
ized for an air strike. General Vandegrift 
ordered ten planes to remain in contin- 
uous support and placed another squad- 
ron on call to Edson. 

By 1045 the resistance had stiffened, and 
the raiders requested that more troops 
land to the west of the village and support 
their attack. Not wanting to weaken the 
perimeter, division replied that such a 
move was not feasible. Vandegrift sug- 
gested that the raiding force reembark and 
return to the perimeter if the Japanese 
proved too strong to handle. Edson re- 
mained, however, and 45 minutes later had 
overrim more artillery pieces as the bat- 
talion advanced slowly against a heavy 
volume of fire. The colonel estimated the 
enemy as about 1,000 well-armed and well- 
equipped troops, and the force now seemed 
inclined to make a stand. Portions of Ed- 
son's advance drew fire from field artillery 
at point-blank range. 

Some of the raider units had lost in- 
ternal contact during the stiffening battle, 
but these faults were corrected at about 
1100, approximately the same time that 

° There was a shortage of landing craft and the 
parachute battalion would not leave the perimeter 
until shortly after 0800. 

the parachute battalion reported to Edson, 
and the commander decided to make a co- 
ordinated attack against the firm opposi- 
tion. The colonel called in a P^OO straf- 
ing attack and then followed this with an 
envelopment inland by his raiders while 
the parachutists protected his flank and 
rear. The assault carried the village, but 
again the Japanese had elected to break 
contact and prepare for an attack at a time 
and place of their own choosing. 

The village was deserted, but the ap- 
pearance of the abandoned encampment 
indicated that reports of native scouts had 
been most accurate. Edson estimated that 
some 4,000 Japanese had been in the vicin- 
ity until shortly before his attack, that 
his force had met only outposts and rear 
guards of a newly arrived unit which ob- 
viously was preparing a strong attack on 
Henderson Field. Twenty-seven Japa- 
nese had been killed. Marine casualties 
numbered two dead and six wounded. 

Edson^ estimate . jai _ ^le-^ Japanese 
strength was a little low, but he was right 
about the enemy's intentions. Just as the 
1st Marines had previously scouted ele- 
ments of the Ichiki Force it later met at 
the mouth of the Ilu River, so Colonel 
Edson had located the gathering Kawu- 
guchi Force his men would meet later in 
a bitter stand before the airfield. 

Rabaul had kept Admiral Tanaka's re- 
inforcement ships busy. The admiral had 
taken over the cruiser Kinugasa to replace 
his damaged Jintsu^ and early on 29 Au- 
gust Admiral Mikawa had ordered Ta- 
naka to begin transporting reinforcements 
by destroyer. The remnants of Ichiki's 
rear echelon would be taken down to Guad- 
alcanal as would the Kawaguchi Force^ 
due to arrive later that day from Truk on 
board the transport Sado Maru. Tanaka 
loaded supplies on board the ships of De- 





stroyer Division 2Ii.^ and put the troops of 
Ichiki's rear echelon on board two de- 
stroyer transports. Then he stood by for 
the arrival of Kawaguchi. 

Kawaguchi's 36th Brigade, a part of the 
18th Division in China, was built around 
Colonel Akinosuka Oka's 12I^th Infantry 
Regiment. From China the unit had 
moved in December 1941 to Borneo. In 
March 1942 it moved to Cebu in the Philip- 
pines, in April to Mindanao, and in June 
to Palau. Alerted for a New Guinea 
operation that never came off, the force 
remained in the Palau Islands until late in 
August when it began to stage in echelons 
through Truk for the Rabaul area. When 
it arrived for this new mission it was 
formed up to include the rear echelon of 
the 2d Battalion, 28th Infantry {Ichihi 
Force), the 12I^th Infantry, the 2d Bat- 
talion, J^th Infantry, and units of artillery, 
engineer, signal, and antitank troops. In 
that form the Kawaguchi Force numbered 
more than 6,000 men.^ 

Admiral Tanaka had his destroyers all 
ready when Kawaguchi arrived. The ad- 
miral met immediately with the general 
to hurry things along, but he ran into dif- 
ficulty at once. Kawaguchi was a barge 
man, and he did not care much for this 
idea of going down to Gu/idalcanal in de- 
stroyers. He had once moved his unit 500 
miles by barges to make a distinguished 
landing on Borneo. Now he wanted to 
know how it would be if he went on down 
to Gizo Harbor just north of New Georgia 
on board his Sado Maru, and then trans- 
ferred to barges for the remainder of the 
trip and for the landing at Guadalcanal. 
Kawaguchi's subordinate officers nodded 
agreement to this idea. They were barge 

men, too. The impatient Tanaka referred 
this dispute to Mikawa of the Eighth Fleet 
and Hyakutake of the Seventeenth Army. 
These officers prevailed upon Kawaguchi 
to temporarily curb his warm regard for 
barges. He would make all of the trip on 
Tanaka's destroyers, and land on Guadal- 
canal from them, besides.^ 

For the build-up on Guadalcanal, Kawa- 
guchi split his command. The general 
would land in the Tasimboko area with 
the Ichiki rear echelon and the 1st and 3d 
Battalions, Infantry Regiment. 
Colonel Oka would land with the remain- 
der of the force — the 2d Battalion, 12J^th 
Infantry — west of Lunga Point near Ko- 
kumbona. Each of the two forces was 
reinforced by a share of the artillery, 
engineers, and other special troops. There 
was only one hitch in the reinforcement 
efforts, even if Kawaguchi might have been 
uneasy without a barge under him, but this 
bobble had no serious over-all results. 
Captain Yonosuke Murakami, command- 
ing Destroyer Squadron 2Jf, was to clear 
the way for the landings by going down 
The Slot on the night of 29 August to at- 
tack a U. S. task force which was reported 
to be off Lunga Point. Instead, Murakami 
came steaming back up The Slot for the 
comfort of the Shortlands. The night sky 
around Guadalcanal, he explained, was too 
full of a bright moon and U. S. aircraft. 
"He was transferred shortly to the home- 
land," Tanaka reported later.* The ele- 
ments of the Kawaguchi Force landed dur- 
ing the nights of 30 and 31 August, at 
about midnight in both cases. 

In spite of the fact that the commanders 
were separated by a distance of some 30 
miles, Kawaguchi planned a difficult 

' 17th Army Ops, I, cited in Miller, 0-uadalcanal, 


' Tanaka Article, I, 695, 697. 
' Hid., 696. 



maneuver that proposed to strike the 
Lunga perimeter in a three-jawed envel- 
opment from the west, south, and south- 
east. It was to be a coordinated attack 
with air and naval support. To the 
normal problems inherent in such an in- 
volved plan, Kawaguchi imposed upon his 
force the additional task of cutting a trail 
over the steep jungle-covered ridges and 
gorges from the Tasimboko area to a point 
south of Henderson Field. The jungle 
trail, planned as a route which would 
enable the Japanese to escape observation, 
was begun about 2 September by Kawa- 
guchi's engineers. Infantry, artillery, and 
other units followed the engineers along 
this hand-hewn jungle route toward their 
lines of departure for the attack against 
the Marines. 


Kawaguchi's fade-out into the jungle 
was successful. He was not spotted by 
Marines again until he was ready to at- 
tack, but it soon became apparent to the 
Lunga defenders that he would have im- 
posing support from Rabaul. Far-rang- 
ing intelligence sources reported a Japa- 
nese naval build-up in the Truk and Palau 
areas and greatly increased air activity 
around the Bismarck Archipelago. 

"The situation as I view it is very 
critical," Admiral Ghormley messaged 
Nimitz.® "Our transportation problem in- 
creases steadily as Japs perfect their 
blockade methods." Japanese pounding 
of Guadalcanal picked up; the defenders 
clearly were being softened up for a big 
attack, and while the South Pacific scur- 
ried to get them more planes the men at 
Lunga hoped that the field would stay 
dry for the important day. 

" Quoted in Marine Air History, 88-89. 

On 11 September, the pace of the attacks 
quickened. Twenty-six bombers and eight 
Zeros came over at 1210 to pock the field, 
kill 11 Marines, and wound 17 others, and 
desti'oy one P-^00 parked beside the strip ; 
and a heavy cruiser and two destroyers 
were spotted steaming south about 100 
miles to the northeast. But on the same 
day the Cactus Air Force added to its 
strength. At 1620 a flight of 24 F4F's 
that had been idle since their carrier 
Saratoga had been torpedoed on 31 August 
came up to Henderson from Espiritu 
Santo under the command of Lieutenant 
Commander Leroy C. Simpler. Before 
noon the next day (12 September) Simp- 
ler's men got their chance to learn Cactus 
operations. Twenty-one of them went up 
with 11 "old" Cactus fliers to shoot down 
12 bombers and three fighters out of a 
42-plane Japanese strike that came over 
at 1100. 

Meanwhile patrols from the 2d Bat- 
talion, 1st Marines began to encounter 
frequent opposition east and southeast of 
the perimeter. Native scouts brought, 
word of large bodies of troops that clearly 
were not wandering remnants of Ichiki's 
action. The troops had an air of purpose 
and direction apparent even to the local 
natives who began to flee from their vil- 
lages to the Marine perimeter. By 10 
September native reports indicated that 
the enemy was less than five miles east of 
the perimeter and that he was cutting a 
road to the south. 

The perimeter by this time had been 
improved and strengthened. The 1st Ma- 
rines right (east) flank was refused for 
some 4,000 yards inland from the mouth of 
the Ilu, and on the west the 5th Marines, 
with a strong reserve in the form of its 2d 
Battalion just over from Tulagi, refused 



its flank inland for approximately half 
that distance. The space inland between 
these flanks still posed a serious problem, 
but it had been partially solved by the 
establishment of well-prepared strong 
points and outposts. (See Map 18) 

Troops from the 1st Amphibian Tractor 
and Pioneer Battalions maintained posi- 
tions south (inland) of the 5th Marines- 
sector west of the Lunga, while east of the 
Lunga a 4,000-yard outpost line was main- 
tained by the 1st Marines, artillerymen, 
the engineer battalion, the bulk of the pio- 
neer battalion, and the raider-parachute 
battalion. General Vandegrift had or- 
dered the raiders and parachutists out of 
division reserve to augment this line by 
preparing positions on a long low ridge 
that extended south of Henderson Field 
and parallel to the Lunga Kiver. The 
thousand-yard-long ridge was but a mile 
south of the airfield and, unless well de- 
fended, offered the Japanese an inviting 
avenue of approach to the field. 

The pioneer battalion (minus its com- 
pany west of the Lunga) held positions 
just south of Henderson Field between 
the Lunga and the north spur of the ridge 
occupied by Edson's force. Farther to the 
east — and across the ridge spur from the 
pioneers — was the area of the engineer 
battalion. Between the two positions was 
the division command post which recently 
had been moved from its former, bomb- 
pocked position near the airfield. 

On the 12th, the same day the Saratoga 
fliers went into business with the Cactus 
circus, Edson and his executive officers 
walked out on their ridge to decide on a 
location for defenses. The officers drew 
small-arms fire from the jungles to the 
south, and Edson called up his troops to 
dig in across the southernmost knoll on the 

ridge. This was forward of the flanks of 
engineers on his left (east) and the pio- 
neers on his right, but Edson wanted to 
hold all the ground he could and to launch 
an attack against the enemy the next day. 

At about 2100 that night a Japanese 
light cruiser and three destroyers entered 
Sealark Channel to shell the airfield, and 
at about the same time the enemy ground 
force probed lightly at the raider-para- 
chute force on the ridge. Fighting was 
sporadic all along the line, and although 
one desultory Japanese attack actually 
made a slight penetration of the Marine 
line, the enemy made no attempt to con- 
solidate or expand this gain. 

Early the next morning (the 13th) Ed- 
son launched his counteroffensive, but he 
found the enemy too strong and well- 
prepared to be thrown back. In the after- 
noon the Marine officer withdrew his ex- 
hausted men north of the positions they 
had held the previous night and estab- 
lished a stronger line on a higher portion 
of the ridge closer to the engineers and 
pioneers to his left and right rear. On the 
right, in the jungle between the ridge and 
the Lunga, a sketchy contact was made 
with the pioneer battalion; on the left 
(east) the raider-parachute flank dangled 
open. (See Map 19) 

While Edson's force sweated under the 
hot sun o\\ the grassy ridge, Henderson 
Field was having more than its share of 
action. The Japanese raids started at 
0950, came back at 1300 and again at 1730. 
The mixed Guadalcanal force shot down 
11 of the enemy planes during the day 
while losing five of their own number. But 
again the U. S. air strength grew. 

Navy pilots from the Hornet and Wasp 
brought in 18 F4F's, and in the afternoon 
more Saratoga fliers and planes came up 



Henderson Field' 
1700 YordI"" 

// : 

// i 
// - 

II - 

II i 


*'»„.. \ 


II : 


" ^ 

II \ 



^]7/7/ ~ Moin enemy thrust 

< j^ I iLJ -Morine front lines 

I I -Commond posts 

100 200 300 400 500 


v...,.v' ,, \ ''H '^X .J„,,o 

MAP 19 




from Espiritu Santo. Nimitz and Ghorm- 
ley were doing all they could to bolster 
the Solomons toe hold against the Japa- 
nese attack that was coming. Later in the 
day Lieutenant Commander Louis J. Kirn 
brought in a flight of 12 SBD's of VS-3 
and the field also got its first torpedo 
planes when Lieutenant Harold H. Larsen, 
USN, flew in leading six TBF's of VT-S.^" 
But while the Henderson flying force 
gained by 60 planes during the period of 
11-13 September, Kabaul's air power 
jumped an additional 140 planes on 12 
September alone. 

Taking periodic cover from sniping and 
bombing raids, Edson's men continued to 
dig in for one more night on the ridge ; on 
the morning of the 14th they were to be 
relieved by the 2d Battalion, 5th Marines. 
But it looked as if the night would be the 
worst they had seen yet; scouting planes 
spotted seven destroyers coming down The 
Slot, evidently to add their bombardment 
to the ground attack that appeared 
shaping up in the jungle to the south. 

During the afternoon the reserve bat- 
talion (2/5) moved to an assembly area 
east of the Lunga and between the airfield 
and Edson's Ridge, and officers of this 
battalion had gone forward to Edson's 
lines to look over the area they would con- 
trol the following day. The 105mm how- 
itzers of the 5th Battalion, 11th Marines 
lay in direct support of the Edson force, 
and elements of the special weapons bat- 
talion had an observation post on the ridge. 
The Guadalcanal defense was as ready as 
it could get. 

'"After this day account-keeping pretty well 
broke down. Records defy determination of who 
flew from Cactus on any given day. In the press 
of fighting, often with "staff" officers in the air 
as much as anybody else, administration and 
office work were marked by extreme casualness. 

Edson's disposition placed his two para- 
chute companies on the exposed left flank 
and tied them in on the right with raider 
Company B which held the ridge knoll in 
the center of the Marine line. Company 
A of the raiders extended down the west 
slope of the ridge toward the Lunga and to 
the makeshift contact with the pioneers. 
Raider Company C, on a high knoll to the 
north (rear) of Company B, was Edson's 

At sunset units were organized in small 
combat groups of about platoon strength 
disposed at intervals along the main line 
of resistance. There were open fields of 
fire only in the center of the position where 
the MLR crossed the grassy ridge, but even 
here the abrupt slopes and broken ground 
made coordination of fires difficult. In 
the last hours of daylight the troops im- 
proved their foxholes and the fields of fire, 
but the resulting positions were neither 
continuous nor complete. 

In the first hours of darkness, Louie the 
Louse, or Washing-Machine Charley," 
chugged over to drop his inconsistent 
scattering of bombs, and about 2100 he let 
go a flare that hung over the field as a 
registration point for the destroyer task 
force that now opened up from Sealark 

As if in answer, a flare went up from 
the troops south of Edson, and without 
artillery preparation Kawaguchi drove a 
two-battalion attack against the center and 
right of the raider-parachute line. Com- 

" Familiar but unaffectionate names by which 
Guadalcanal defenders identified the nuisance 
raiders that droned around almost nightly. Tech- 
nically, "Charlie" was a twin-engine night bomber 
from Rabaul, "Louie" a cruiser fioat plane who 
signalled to the bombardment ships. But the 
harassed Marines used the names interchange- 



pany B's central sector on the high knoll 
caught most of this first assault and turned 
it back, but the other attack column found 
an opening to the west and came through 
to cut off and envelop Company B's right 
platoon. While the Japanese drove 
through this gap between Companies A 
and B, the isolated platoon fought its way 
back along 250 yards of the ridge to join 
Company C on the knoll to the north. 
Still engaged and nearly overpowered, 
Company B refused its right flank along 
the ridge's west slopes. (See Map 20) 

Edson had been calling in fire from 
5/11's howitzers since the beginning of 
the attack, and as the Japanese continued 
to hammer at his men the colonel directed 
the artillery closer and closer until it was 
falling within 200 yards of the Company 
B lines. But still the Japanese came on, 
and by 2200 Edson estimated that the two 
understrength parachute companies and 
Company B (less the withdrawn platoon) 
were opposed by at least two enemy bat- 
talions attacking in full force. 

Japanese infiltration parties were taking 
over some of the Company B foxholes, 
communication lines were cut throughout 
the area, and the Japanese now began to 
drum the ridge with heavy mortar fire. 
Following a violent barrage at 2230, the 
Japanese attack shifted to the east where 
it struck the thin flank held by the para- 
chute troops. Screaming in English, "Gas 
attack ! Gas attack !", the Japanese came 
out of the jungle through a smoke screen 
and drove the parachutists back along 
the ridge to expose the left flank of Com- 
pany B. 

This left the B Company raiders, now 
cut to approximately &0 men, exposed on 
both flanks as well as their front, and 
Edson called for them to pull back to a 
last-ditch stand with Company C. Com- 

pany A would join the force there, and 
Edson ordered his men to hold at all costs. 
It was the last dominating terrain feature 
south of the airfield. 

Screening the withdrawal of the two 
companies with artillery fire, Edson col- 
lected his men as they filtered back and 
built them up in what he hoped would be a 
line strong enough to make the final stand. 
The colonel and his officers ironed out the 
confusion of setting in the new defense in 
darkness and under fire while holding off 
repeated Japanese assaults. In all, the 
enemy struck more than a dozen times 
throughout the night, the Kawaguchi men 
grinding themselves into the fire from 
Marine artillery, mortars, machine guns, 
and rifles in vain attempts to dislodge 
Edson from his final knoll of Bloody 
Ridge.^^ Japanese flares "telegraphed" 
each attack, providing the 11th Marines 
gunners with reference points for their all- 
night firing in which they expended 1,992 
rounds of 105mm projectiles, some at 
ranges as short at 1,600 yards. 

At 0400, with the Japanese attacks still 
in progress, companies of reserve battalion 
2/5 began to move singly through the 
darkness and into positions on the raider- 
parachute left flank. Darkness and un- 
certainty about Edson's new location 
brought confusion to this reinforcement 
effort, but the companies succeeded in 
gaining positions from which they aided 
in standing off the final Japanese attacks. 

While the action on the ridge was in 
progress, another Japanese unit (possibly 
the Ichiki rear echelon) struck farther to 
the east where the right flank of the 3d 
Battalion, 1st Marines lay exposed near 
the Ilu River inland. Striking: with a 

" The name, used interchangeably with "Ed- 
son's Ridge," was employed after the battle, to 
identify this terrain feature. 





force of about two reinforced companies, 
the Japanese engaged the Marines in a 
night-long fire fight but failed to penetrate 
the line. 

In another, lesser action of the night a 
patrol of some 30 Japanese, evidently from 
the force that penetrated Edson, wandered 
into a thin line of Company C, 1st Engi- 
neers in the area east of the division CP 
and near 5/11's Headquarters and Service 
Battery south of the airfield. The line 
had been thinned earlier in the night when 
Company A of the engineers had been 
called back to aid the CP defense, and the 
Japanese patrol which struck at 0530 suc- 
ceeded in taking two left flank machine- 
gun positions before headquarters and 
service artillerymen came up to bolster the 
line and help evacuate wounded. The 
Japanese heckled the line for the shprt 
time remaining until daylight, then retired 
into the jungle. Four engineers were 
killed and 14 were wounded. Ten Japa- 
nese bodies were buried in the area. 

Also by daylight (14 September) the 
attacks on Edson's Ridge and the 3/1 line 
had dwindled to sporadic sniping, and in 
the Edson Ridge sector the disorganized 
Japanese were bombed and strafed into 
retreat by three P-400"s from Henderson 
Field. Survivors remaining near the 
ridge were hunted down and killed. 

After this the only enemy still in action 
was a force of about battalion strength 
which fired across the Ilu plain some dis- 
tance east of Bloody Ridge and harassed 
the Marines of 3/1 (Lieutenant Colonel 
William N. McKelvy, Jr.) who held that 
portion of the Marine line. Tanks were 
called up against this enemy force, and 
after a hasty reconnaissance six of these 
vehicles moved forward without infantry 
support toward the Japanese line in the 
fringe of jungle by the Tenaru. Two 

tanks were hit almost at once by a Japanese 
antitank gun. Another tank charged 
across the plain and over a grass hut only 
to plunge down a 30- foot bank into the 
Tenaru ; all four crew members were killed. 
A fourth tank was hit by this antitank gun 
shortly after this, the fifth tank returned 
to the infantry lines, and the sixth tank 
was stopped by a wrecked track 50 yards 
in front of the Japanese gun. The men in 
this tank bailed out and returned to the 
infantry position. The tank attack had to 
be chalked off as a costly failure, but the 
Japanese caused little trouble in the area 
after this. A desultory fire fight con- 
tinued across the plain until 16 September 
when the enemy withdrew. 

Tactically the entire Kawaguchi Force 
could be scratched. About 400 of the 
Ichiki rear echelon subsequently reached 
Koli Point as did some troops of the 2d 
Battalion^ l^th Injamtry^ but these were 
hardly more than stragglers. The remain- 
der of the force — the larger element which 
had struck Edson's Ridge — reduced itself 
to a rabble while cutting a tortuous jungle 
trail over the southern slopes of Mount 
Austen, across the up-country Matanikau 
territory, and finally to Kokumbona. 
Wounded died along the route and equip- 
ment was abandoned by the weakened, ex- 
hausted survivors. 

The Marines had turned back a serious 
threat to their precarious Guadalcanal po- 
sition, but again a part of the thanks could 
go to Japanese bungling — on the battle- 
field as well as in planning at higher eche- 
lons. Although Kawaguchi salvaged 
enough pride to spare himself the hara- 
kiri fate of Colonel Ichiki, he still was 
only a slightly stronger boy whom Tokyo 
and Rabaul hopefully had sent away on 
a man's job. 



RAIDERS' RIDGE looks calm in this after-action shot, hut it was the scene of a violent and bloody fight 
crucial to the defense of the perimeter. ( USMC 50007 ) 

MARINES OF THE 2D RAIDER BATTALION land at Aola Bay, starting point of their month-long 
operations behind the enemy lines east of the perimeter. (USMC 51359 ) 


Action Along the Matanikau 

Retreat of the Katoaguchi Force prom- 
ised the Marines of the Lunga perimeter 
another breathing spell from ground at- 
tacks, but there was no time for relaxation 
or relief from concern about the future. 
Air and naval strikes continued to pound 
the Henderson Field defenders, and aerial 
reports of a continued Japanese build-up 
at Rabaul forecast additional attempts to 
retake the Guadalcanal area. Patrolling 
schedules were stepped up ; it was disquiet- 
ing to know that both the Ichiki and 
Kamaguchi Forces had landed on the 
island and moved into attack positions 
without the Marines once being completely 
sure of their exact locations. 

At the conclusion of the Battle of the 
Ridge on 14 September, the Marines had 
been ashore for 38 days without receiving 
either reinforcements or additional ammu- 
nition. For most of this period the men 
could be fed only two meals a day, and part 
of this food came from captured Japa- 
nese supplies. Malaria was beginning to 
add its toll to battle casualties, and al- 
though defensive emplacements were con- 
tinually improved, the Marine force was 
wearing itself down while the Japanese 
ground strength continued to mount in 
staging areas in the Bismarcks. 

In these lean early days in the Pacific, 
the problem of new strength for the 
Guadalcanal effort was a thorny one. The 
Solomon Islands position was merely a 
salient, and still not a strong one, which 
made a questionable contribution to the 

safety of other Allied positions farther to 
the south. So these areas could not be 
stripped of defenders, and even if some 
spare troops could be found there still was 
another operation slated. From the first, 
the plan for this initial Allied offensive 
in the Pacific had included an occupation 
of Ndeni Island in the Santa Cruz group 
southeast of the Solomons. 

The 2d Marines first had been sched- 
uled for this job, but Vandegrift had been 
allowed to keep this regiment when the 
opposition became so bitter on Tulagi. 
Later the general requested that his di- 
vision's third organic infantry regiment, 
the 7th Marines, come over from its 
Samoan garrison duty with its support- 
ing artillery, the 1st Battalion, 11th Ma- 
rines. But Admiral Turner demurred ; he 
still saw a need for the Ndeni operation, 
and the reinforced 7th Marines was the 
only amphibious force readily available 
for such an undertaking. On 20 August 
the admiral published his Ndeni plan, and 
on 4 September the 7th Marines with its 
artillery and part of the 5th Defense 
Battalion sailed from Samoa for Espiritu 

' Turner and Vandegrift often disagreed on 
conduct of the Guadalcanal operation ashore and 
on progress of the Solomon Islands action in 
general. Turner found fault with Vandegrift's 
perimeter concept of defense. His idea was to 
disperse the Marines along the Guadalcanal coast 
and set them upon the task of mopping up the 
remaining Japanese. 




But by 9 September, with the 7th 
Marines' convoy still en route, Turner 
agreed with Vandegrift's August request 
for control of this infantry regiment, and 
he requested Admiral Ghormley's permis- 
sion to divert the regiment from the Ndeni 
operation. The issue still was not won 
for the Marine general, however. Turner 
believed this fresh unit should set up 
coastal strong points outside the Lunga 
perimeter, while Vandegrift held that a 
reinforcement of his perimeter was the 
more pressing need. Turner relayed this 
question to Ghormley on 12 September, 
the same day the 7th Marines arrived in 
the New Hebrides, and Ghormley next day 
ordered the reinforced regiment to move 
as soon as possible to the Guadalcanal 

After unloading the 5th Defense Bat- 
talion units at Espiritu Santo, the ships 
bearing the infantry regiment and the 
artillery battalion departed for Guadal- 
canal on 14 September, the same day that 
the 3d Battalion, 2d Marines was brought 
across Sealark Channel from Tulagi. 
Operating with three cruisers plus the 
destroyers and mine sweepers of the newly- 
formed Task Force 65, the transports spent 
four days at sea skirting enemy naval 
forces in the Solomons waters. The con- 
voy finally anchored off Kukum early in 
the morning of 18 September. 

The trip cost the Navy dearly. Carriers 
Hornet and Wasp.^ then the only flattops 
operational in the entire South Pacific 
(both the Saratoga and the Enterprise 
were under repair) ranged southeast of the 
Solomons with other escort support for 
the convoy, and the Japanese had just 
sown the area with a division of sub- 
marines. The Wa^p caught two torpe- 
does, burned and sank; the battleship 

North Carolina was damaged as was the 
destroyer O^Brien, which later broke in 
two and went down while heading back 
to the U. S. following temporary repairs. 

But for Henderson Field there was ad- 
vantage even from such grim disasters as 
this; pilots and planes that otherwise 
would have been flying from their carriers 
could now come up to give the Cactus 
Force a hand. On 18 September six Navy 
TBF's arrived in the Lunga area, and on 
28 September 10 more planes, some SBD's 
and the other TBF's, flew in. Although 
enemy raids dropped off somewhat after 
the defeat of the KawagucM Force, opera- 
tional losses still drained Geiger's air 
power, and such reinforcement managed 
only to keep the Cactus Force at a 50-to- 
70-plane level, but for this Lunga was most 

September 18th was a red-letter day for 
the Guadalcanal defenders. While the 
reinforced 7th Marines unloaded its 4,262 
men, three other tra nsports which were not 
part of TF 65 entered the channel with an 
emergency shipment of aviation gasoline. 
In all, this shipping put ashore 3,823 
drums of fuel, 147 vehicles, 1,012 tons of 
rations, 90 per cent of the 7th Marines sup- 
plies of engineering equipment, 82.5 per 
cent of the organizational equipment, and 
nearly all of the ammunition.^ Turner's 
force then took on board the 1st Parachute 
Battalion, 162 American wounded, and 
eight Japanese prisoners and departed for 
Espiritu Santo at 1800. 

After this successful unloading, men on 
Guadalcanal began to draw more adequate 

^ This was the first ammunition Marines had 
received since the landing. It included about 10 
units of Are with additional hand grenades and 
81mm mortar shells. FinalRept, Phase V, Annex 
Z, 10. 



THE PAGODA AT HENDERSON FIELD, headquarters of Cactus Air Force flyers throughout the first 
months of operations from the captured airfield. ( USMC 50921 ) 

CACTUS AIR FORCE spreads in all its variety across Henderson Field during a lull in the battle; in the 
foreground are Marine scout-bombers. ( USMC 108580) 



rations, and General Vandegrift was able 
to adopt new defensive concepts for his 
force of some 19,200 men now at Lunga. 
Local air power made a counterlanding 
less likely, and the attack pattern set by 
Ichiki and Kawaguchi indicated that more 
attention should be given to the inland rim 
of the perimeter. On 19 September, 
Vandegrift's Operational Plan 11-42 pro- 
vided for this new concept by dividing the 
defenses into new sectors with increased 
all-around strength. 

Relieving special troops such as the 
engineers and pioneers, infantry battalions 
filled the yawning gaps that previously 
had existed south of the airfield and along 
the southern portions of the new inland 
sectors. The pioneers, engineers, and the 
amphibian tractor personnel now were 
able to perform their normal functions 
during the daylight hours and at night 
bolster the beach defenses where fewer 
men were needed. Each infantry regi- 
ment maintained a one-battalion reserve, 
one or all of which could be made available 
as a division reserve if necessary.^ 

Gaps still existed in the perimeter. 
Generally the lines followed the' high 
ground of the ridges, but intervening 
stretches of low jungle often could not be 
occupied in mutually supporting positions. 
Barbed wire had become available ir^ in- 

'All large trucks were called in to a division 
pool each night to stand by in case a division 
reserve had to be trucked quickly into action. 
Although many additional vehicles had been 
landed on the day the 7th Marines arrived, the 
division still was critically short of motor trans- 
port. The supply of large trucks never topped a 
bare 30 per cent of the allowance. In rear areas 
there was a mistaken idea that one-and-a-half 
ton and larger trucks could not be used on 
Guadalcanal. IMd., 1. 

creasing quantity, and in most sectors 
double apron fences stretched across the 
ground in front of infantry positions of 
foxholes and logged and sand-bagged ma- 
chine-gun emplacements. Colonel Robert 
H. Pepper's 3d Defense Battalion, with 
the 1st Special Weapons Battalion at- 
tached, retained responsibility for antiair- 
craft and beach defense, and Colonel 
Pedro A. del Valle's 11th Marines, bol- 
stered by its 1st Battalion, remained in a 
central position supporting all sectors. 

The 1st Marines retained responsibility 
for the east side of the perimeter, from an 
area near the mouth of the Ilu River in- 
land to a point beyond the former right 
flank where McKelvy's battalion had 
fought the Japanese across the grassy 
plain. The fresh troops of the 7th Ma- 
rines joined the 1st Marines at that point 
and extended across Edson's Ridge to the 
Lunga River. Beyond that river the 3d 
Battalion, 2d Marines built up a line that 
tied in on the right to the positions of the 
5th Marines, and this latter regiment 
closed the perimeter with its right flank 
which connected with the left flank of the 
3d Defense Battalion at the beach. 

Tentative plans in the reorganization 
also included extending the perimeter with 
strong points of one- or two-battalion 
strength to the mouth of the Matanikau on 
the west and the Tenaru on the east. Such 
positions would take advantage of the nat- 
ural defensive potential of the two rivers 
and aid the Marines in blocking Japanese 
movements in strength toward the main 
battle positions. These strong outposts 
were not established at this time, however. 
(See Map 21) 

The first order of business seemed to re- 
quire aggressive attention to the west. 



Patrol actions had confirmed intelligence 
estimates that a strong enemy force was 
operating from the Matanikau village area 
on the west bank of the river, but that 
from the southeast or east there seemed 
little danger of an attack. With the Hen- 
derson Field side of the Lunga perimeter 
thus reasonably safe from an attack in 
force, the division planned a series of ac- 
tions to clear the Matanikau sector. Jap- 
anese troops there included elements from 
the Ifth Infantry Regiment of the 2d Dwi- 

sion, and other personnel of the Kawa- 
guchi Force. The Infantry had been 
reinforced by new Japanese landings of 

The first action against this enemy force 
sent Lieutenant Colonel I^ewis B. Puller's 
1st Battalion, 7th Marines into the Mount 
Austen area on 23 September. The Ma- 
rines were to cross the Matanikau upstream 
and patrol between that river and the vil- 
lage of Kokumbona. The action was to be 
completed by 26 September at which time 



the 1st Raider Battalion * was to advance 
along the coast to Kokumbona where a 
permanent patrol base was to be estab- 

After passing through the perimeter on 
23 September, Puller's battalion next day 
surprised a Japanese force bivouacked on 
the Mount Austen slopes, and scattered the 
enemy in a brief clash that ended shortly 
after nightfall. The action cost Puller 7 
killed and 25 wounded, and the commander 
requested air support for a continuation 
of his attack the next day (25 September) 
and stretchers for 18 of his wounded men. 

Realizing that a prompt evacuation of 
18 stretcher cases over the rugged terrain 
would take at least 100 able-bodied men, 
General Vandegrift sent Lieutenant 
Colonel Rosecrans' 2d Battalion, 5th Ma- 
rines out to reinforce Puller. With this 
new strength to back him up, Puller sent 
a two-company carrying and security force 
back with the wounded and pushed on 
toward the Matanikau. 

The general's 24 September communica- 
tions with Puller also gave the colonel 
the prerogative of altering the original 
patrol plan so that he could conform to 
the termination date of 26 September. 
Accordingly, when 1/7 and 2/5 reached the 
Matanikau on 26 September they did not 
cross but patrolled northward along the 
east bank toward the coast. 

At about 1400 the two battalions reached 
the mouth of the river and there began to 
draw fire from strong Japanese positions 
in ridges on the west bank. Companies E 
and G of 2/5 attempted to force a crossing 
but were repulsed, and soon were pinned 

* Now commanded by LtCol Samuel B. Griffith, 
II. Edson, the former commander, had recently 
advanced to the rank of colonel to take over the 
5th Marines on 21 September. Edson succeeded 
Col Leroy P. Hunt who had departed for the U. S. 

down by fire from automatic weapons. 
Puller called in artillery and air, but the 
enemy positions remained active. By 1600 
the combined forces of Puller and Rose- 
crans had sustained 25 casualties, and the 
action ^was broken off while the Marines 
strengthened their positions for the night. 

Meanwhile the raider battalion, on its 
way to establish the patrol base at 
Kokumbona, had reached the vicinity of 
the fire fight, and division directed Grif- 
fith to join with 1/7 and 2/5 and to pre- 
pare for a renewal of the attack next day. 
With this large provisional group now 
formed, Vandegrift sent Colonel Edson 
up to take command. Puller would act as 
executive officer. Edson's plan for the 
coordinated attack next day (27 Septem- 
ber) called for the raiders to move some 
2,000 yards inland, cross the Matanikau, 
and envelop the enemy right and rear 
while 1/7 supported by fire and 2/5 struck 
frontally across the river near its mouth. 

The attack began early on 27 Septem- 
ber, but failed to gain. Marines of 2/5 
could not force a crossing, and the raiders' 
inland maneuver stopped short when Grif- 
fith's battalion encountered a Japanese 
force which had crossed the river during 
the night to set up strong positions on high 
ground some 1,500 yards south of the 
beach. First fire from mortars and auto- 
matic weapons wounded Griffith and killed 
his executive officer, Major Kenneth D. 
Bailey, one of the heroes of the Battle of 
the Ridge. 

A raider message reporting this action 
unfortunately was confusing, and from 
it Edson concluded that the battalion had 
succeeded in gaining the enemy right flank 
beyond the river and that the fight was in 
progress there. Thus misinformed, the 
colonel ordered the raider battalion and 



2/5 to resume their attacks at 1330 while 
the 1st Battalion, 7th Marines (less Com- 
pany C) made an amphibious envelopment 
west of Point Cruz to strike the Japanese 
Matanikau line, from the rear. 

Under the command of Major Otho 
Rogers, the 1/7 troops left Kukum in land- 
ing craft just as a strong bombing raid 
came over from Rabaul. The division 
command post took heavy hits which 
wrecked communications, and the de- 
stroyer Ballard^ supporting the landing, 
had to slight her mission while taking 
evasive action. The landing at 1300 was 
unopposed, however, and the companies 
pushed rapidly inland toward a high 
grassy ridge about 500 yards from the 

But as the leading elements reached the 
top of this ridge, they were taken under 
mortar and small- arms fire. Major 
Rogers was killed by a mortar round and 
Captain Zach D. Cox, Company B com- 
mander, was wounded. Captain Charles 
W. Kelly, Jr., acting second in command, 
took charge of the battalion just as the 
enemy cut the Marines off from the beach. 
Kelly found that he could not communi- 
cate with the perimeter, and the close-in 
fight with the surrounding enemy grew 
rapidly more desperate. The Company 
D mortar platoon had only one of its 
weapons and about 50 rounds of ammuni- 
tion, and to bring this weapon to bear on 
the pressing Japanese a mortarman had 
to lie on his back with his feet supporting 
the nearly-perpendicular tube from the 
rear while Master Sergeant Roy Fowel 
called the range down to 200 yards. 

Fortunately, Second Lieutenant Dale 
M. Leslie flew over at about that time in 
his SBD. As pilot of a plane incapable of 
dogfighting in the bomber pack, Leslie was 

hunting likely targets while staying clear 
of the field and air engagements. As he 
circled overhead, the Marines below 
spelled out the word "Help" in white 
undershirts laid on the hillside, and Leslie 
managed to make radio contact with Ed- 
son at the mouth of the Matanikau and 
relay this distress signal. 

That was summons aplenty for Puller, 
chafing in Edson's provisional command 
post while his battalion went off to battle 
without him. The combined attack at the 
river mouth and inland clearly had mis- 
carried, and his men in 1/7 stood exposed 
to the full wrath of the Japanese west of 
the river. AVith characteristic directness, 
the lieutenant colonel collected the landing 
craft and churned out to board the Bal- 
lard. The ship and her skipper, soon 
under the Puller spell, steamed to the 
rescue close ashore, the landing craft in the 
wake ready to be used for a withdrawal. 

It was a day for heroic action. When 
the force trapped ashore saw the ship 
coming down the coast, Sergeant Robert 
D. Raysbrook stood out on a hillock of the 
ridge and semaphored for attention. 
From the bridge of the Ballard Puller 
ordered his men to pull out to the beach. 
Raysbrook, still exposed to the enemy fire, 
flagged back the information that their 
withdrawal had been cut off. The ship 
then asked for fire orders, and with Cap- 
tain Kelly relaying his signals through 
the sergeant, batteries on the Ballard 
began to blast out a path to the beach. 

Supporting fire from the ship was a 
deciding factor in the action, but the com- 
panies still had a fight ahead of them. 
Japanese artillery began to take casualties 
as the Marines withdrew fighting thi'ough 
the enemy infantry still pressing from the 
flanks and rear. Platoon Sergeant An- 



thony P. Malanowski, Jr. took a Browning 
automatic rifle from a man dropped in 
action and covered the withdrawal of 
Company A until he himself was overrun 
and killed by the Japanese. But by then 
his company had reached the beach where 
it set up a hasty defense into which Com- 
pany B and elements of Company D drew 
shortly thereafter. 

With the Marines fighting off the enemy 
at their rear, the landing craft now moved 
shoreward to begin their evacuation, and 
thereby exposed themselves to heavy Japa- 
nese fire from the high ground above the 
Marines on the beach and from the pro- 
jecting terrain of Point Cruz to the east. 
The Japanese were determined not to allow 
a thwarting of their trap, and the stiffen- 
ing crossfire drove the craft back offshore 
where they bobbed in ground swells and 

This was observed by Lieutenant Leslie, 
still keeping a watchful eye on the action 
from his SBD, and he came down again 
to lend a hand. The pilot strafed the 
Japanese positions and then turned to 
make a few swooping passes over the land- 
ing craft to herd them on their way. Thus 
heartened and hurried along, the cox- 
swains went back in to the beach. 

The fire from the beach, although damp- 
ened by the strafing SBD, still was heavy, 
but Signalman First Class Douglas A. 
Munro of the Coast Guard, coxswain of 
the craft, led the other coxswains through 
it and maneuvered his Higgins boat to 
shield the others. The Marines loaded on 
board with their wounded while Munro 
covered them with the light machine guns 
on his boat. He ordered his boat away 
when the other craft were clear, and still 
firing, was making his own withdrawal 
when he was killed by fire from the beach. 

The miniature flotilla returned to the 
perimeter landing site at Kukum by night- 
fall. The action had cost this battalion 24 
killed and 23 wounded. The raiders and 
2/5 likewise withdrew after 1/7 got safely 
clear of the Point Cruz area, and their 
casualties added another 36 dead and 77 
wounded to the tally for the operation. 


Costly as this action at the Matanikau 
had been, it confirmed the data being col- 
lected by intelligence agencies, and these 
facts over-all were as important as they 
were disquieting. Japanese ships still 
entered Guadalcanal waters nearly every 
night, barges beached along the coast indi- 
cated many new landings, air attacks had 
picked up again since a comparative lull 
following the Battle of the Ridge, and now 
it was clear that the Japanese troops as- 
sembling on the island were concentrating 
just beyond the Matanikau. Another and 
a stronger Japanese counteroffensive 
loomed, and although defeat of the Ichiki 
and Kawaguchi Forces gave the Marines a 
new confidence in their ability to hold the 
perimeter, there was yet another factor. 
Late in September the Japanese began to 
land 150mm howitzers, and these weapons 
would be capable of firing on Henderson 
Field from the Kokumbona area. 

Cactus fliers continued to hold their own 
against enemy air attacks of the field; 
Japanese gunfire ships had to come late 
and leave early to avoid the U. S. planes in 
daylight encounter, and the frequent night 
raids of Washing-Machine Charlie were 
more damned than damaging. But big 
howitzers were something else. The Ma- 
rines had no weapon that could reach a 
1 50mm in counterbattery , and they had no 
sound-flash equipment to locate such firing 
positions, anyway. If the Japanese could 



add the effective fires of these weapons to 
air raids and naval shelling, it might be 
just enough tip of balance in their favor 
to hold down the Cactus fliers while a large 
force mounted to dislodge the Americans 
from the Lunga. 

Accordingly, an attack was scheduled to 
trap the enemy force and drive survivors 
beyond artillery range, and a success in 
this would be followed by establishment 
of a permanent patrol base at Kokumbona 
which could make sure the long-range field 

pieces stayed out of range. The plan of 
attack was similar to that of the operation 
which had just failed, but this new effort 
would be made in greater strength. The 
5th Marines (less one battalion) would 
engage the enemy at the river mouth while 
the 7th Marines (also less one battalion) 
and the 3d Battalion, 2d Marines, rein- 
forced by the division scout-sniper detach- 
ment, would cross the river inland and then 
attack north toward Point Cruz and 
Matanikau Village. (See Map 22) 



Colonel William J. Whaling,^ who com- 
manded 3/2 and the scout-snipers on this 
special mission, was to lead to envelopment 
by crossing the Matanikau some 2,000 
yards upstream and then attacking north 
into the village on the first ridge west of 
the river. Whaling would be followed by 
the 7th Marines battalions which would 
also attack north abreast and to the left of 
the Whaling group. 

The 1st Marine Aircraft Wing, with its 
composite Cactus Force, was to provide 
planes for infantry liaison, close air sup- 
port, and artillery spotting. In the artil- 
lery plan, 1/11 would support the 7th 
Marines; 2/11, the 5th Marines; 5/11, 
the Whaling Group; and 3/11 would be 
in general support of the Lunga perim- 
eter. If all went well, Whaling's assault 
of Japanese positions near the coast 
would be followed by a 5th Marines river 
crossing, a passage of Whaling's lines, 
and a pursuit of the enemy toward Point 
Cruz where the 7th Marines on Whaling's 
left would close the trap in front to the 
withdrawing enemy. The 3d Battalion, 
1st Marines provided the division reserve 
for the operation," and Vandegrift's com- 

' Whaling had been promoted to the rank of 
colonel shortly after the Guadalcanal landing, 
and although there was no billet for an additional 
colonel in the division, he stayed on to train 
scouts and snipers in practical, combat skills. 
Graduates of the course returned to their outfits 
to pass on their knowledge and thus increase the 
division's general proficiency in patrolling ; 
others replaced these graduates and Whaling's 
schooling continued. As a unit, the scout-snipers 
normally operated independently, but sometimes, 
as in this case, joined other commands during a 
special mission. 

' There was an indication that this unit might 
be employed in an amphibious envelopment. Ar- 
rangements had been made for a boat group, 
and McKelvy's battalion was on a 30-minute 
standby. FinalRept, Phase V, Annex D. 

mand post would coordinate the entire 
operation. Movements of the forces were 
to get underway on 7 October, and the 
coordinated attack would jump off on 8 

From recent experience with this grow- 
ing Japanese force, the Marines expected 
a stiff fight with all the usual and unusual 
obstacles encountered in battle. But in 
this case there was to be one large factor 
they had no reason to suspect. By an un- 
fortunate coincidence the Japanese also 
had set 8 October as the date for an at- 
tack of their own, and their scheme could 
hardly have been a better counter against 
the Marines had they been looking over 
the shoulders of Vandegrift's staff. Ra- 
baul had ordered Colonel Tadamasu 
Nakaguma to cross the Matanikau on 8 
October with his 4-ih Infantry and estab- 
lish artillery positions which could sup- 
port the new counterattack then in plan- 
ning. To accomplish this mission, Naka- 
guma sent an enveloping force inland 
across the Matanikau on 6 October while 
he slipped the cautious first echelon of a 
bridgehead across the river near the coast. 
There the Japanese forces met the Marines 
who moved from the Lunga perimeter at 
0700 on 7 October. 

Whaling's Group scrapped for several 
hours with the inland Japanese force 
which confined its opposition to sniping 
and harassment, but by the middle of the 
afternoon Whaling decided to bypass the 
enemy. At nightfall the envelopment 
force bivouacked on high ground south of 
the Matanikau's fork, the designated as- 
sembly area for the 8 October attack, and 
the Japanese did not pursue. Meanwhile 
the 5th Marines met with greater difficulty 
from Nakaguma's men near the river 



The advance guard of the 3d Battalion, 
5th Marines came under fire from this 
enemy at about 1000, and the battalion de- 
ployed forward in an attack while the 2d 
Battalion swung to the left around the 
action and reached the river without op- 
position. The Japanese gave ground to 
previously prepared positions, but 3/5 was 
unable to push them beyond this line in 
spite of flanking assistance from 2/5. 
Vandegrift reinforced Edson with an un- 
derstrength raider company, but the 
Japanese continued to hold their confined 
bridgehead some 400 yards inland from 
the beach, and the Marines drew up for 
the night. They held a 1,500-yard front 
which extended inland from the coast and 
bowed around the Japanese pocket on the 
river's east bank. During the night the 
5th Marines and some amphibian tractors 
simulated noisy preparations for a tank- 
supported river crossing to divert Japa- 
nese attention from the Wlialing-7th 
Marines envelopment force. 

Heavy rains which began that night, 
and continued into the 8th, made trails 
and hills slick, muddy, and treacherous, 
and grounded the Cactus fliers. The at- 
tack had to be postponed, but the 5th Ma- 
rines and raiders continued to reduce the 
Japanese positions on the east bank. At 
about 1830 the Japanese, under pressure 
all day from the Marines, made a final ef- 
fort to break out of their nearly sur- 
rounded bridgehead and retreat across the 
river mouth. Kunning abreast, the enemy 
troops charged from their foxholes against 
the thinly-held Marine right flank where 
the raiders faced them. Front rank at- 
tackers engaged the Marines with small- 
arms fire while succeeding ranks pitched 
hand grenades into the raider positions. 
Some hand-to-hand fighting resulted, and 

casualties were high on both sides. Twelve 
raiders were killed and 22 wounded, while 
counted enemy dead numbered 59. Some 
of the surviving Japanese managed to 
escape across the river, and the bridgehead 
was completely reduced. 

While the coordinated Marine attack 
waited out the rain, division was warned 
by higher intelligence sources that the 
expected strong Japanese counteroffensive 
appeared close at hand; aerial observers 
and coastwatchers to the north reported 
increased troop activity and a shipping 
concentration around Rabaul. General 
Vandegrift accordingly scaled down his 
planned attack to merely a raid in force 
so that no major troop strength would be 
beyond a day's march of the perimeter. 

This decision did not alter the basic 
envelopment maneuver, however, and on 9 
October Whaling and the 7th Marines 
moved across the Matanikau and attacked 
rapidly northward to raid the Point Cruz 
and Matanikau village areas. Wlialing's 
Group moved along the first high ground 
west of the river ; Lieutenant Colonel Her- 
man H. Hanneken's 2/7 moved north on a 
ridge some 1,000 yards off Whaling's left 
flank, and Puller with 1/7 attacked along 
another ridge west of Hanneken. 

Whaling and Hanneken reached the 
coast without serious opposition while 1/7 
on the extreme left encountered a strong 
force of Japanese in a deep ravine about 
1,500 yards inland from Point Cruz. 
Puller brought artillery and mortar fire 
down on the Japanese, and his men picked 
off the enemy with rifle and machine-gun 
fire as they climbed the far side of the 
ravine to escape the indirect fire. A few 
enemy escaped up the steep slope, but most 
of them were either killed by small-arms 
fire or driven back down the hill into the 
mortar and artillery concentration. 



It was a most effective arrangement for 
methodical extermination, and Puller 
and his men kept it up until mortar ammu- 
nition ran low. Then they withdrew to 
join the Whaling Group and Hanneken, 
and by 1400 the combined raiding force 
had retired east of the Matanikau through 
the covering positions of the 5th Marines 
and the raiders/ The three-day operation 
had cost the Marines 65 dead and 125 
wounded. A Japanese diary found later 
by Marines placed the J^th Infantry losses 
at 700 men. 

' Positions at the river mouth were retained to 
guard against a new Japanese crossing. 

Rain and the threat of a new counter- 
offensive had thwarted the Marines' at- 
tack plans, but the action could still go 
down in the gain column. The raid had 
tripped up the attack Colonel Nakaguma 
had planned for the same period, and it 
had done away with a great number of his 
men. And in the short time that men of 
the 7th Marines had been ashore on the 
island, they had earned a right to identifi- 
cation as veteran troops. So with a com- 
pletely combat-wise division on hand — 
and Army reinforcements on the way — 
Vandegrift and his staff now made plans 
to meet the strong Japanese attack that 
was bearing down upon them. 


Japanese Counteroffensive 

In spite of the miscarriage of Naka- 
guina's effort to establish a bridgehead 
across the Matanikau, the Japanese 
Seventeenth Army continued preparations 
for its big push. On 9 October, the same 
day that Lieutenant Colonel Puller caught 
a major portion of Nakaguma's 4-ih In- 
fantry between the devil of small-arms 
and the deep sea of artillery and mortar 
concentrations, Seventeenth Army General 
Haruyoshi Hyakutake landed on Guadal- 
canal to take personal charge of the 
Japanese campaign. 

Things were serious but not desperate. 
Although Ichiki and Kawaguchi had al- 
lowed unfounded optimism and overcon- 
fidence to swamp their missions against 
the Marines, Hyakutake still had a strong 
force and a proud confidence that he could 
wipe out the Lunga positions in one blow. 
And with Guadalcanal safely back in 
Japanese hands, Imperial troops then 
would retake Tulagi and occupy Rennell 
and San Cristobal. At the same time 
Seventeenth Army reserves and the Jap- 
anese Navy could renew attacks in New 
Guinea and take Port Moresby by late 
November. The Bushido spirit would be 
back at full strength. 

By early October the Japanese had 
brought troops in from the Philippines, 
the East Indies, China, and Truk to place 
within the Seventeenth Army command in 
Rabaul and the Solomons two divisions, a 
brigade, and a reinforced battalion. Sup- 
port forces included six antiaircraft bat- 

talions plus one other AAA battery; a 
heavy regiment and an independent tank 
company ; one regiment and one battalion 
of mountain artillery; an engineer regi- 
ment, and other troops including a mortar 
battalion and a unit of reconnaissance air- 
craft. Included in this general listing 
were the Kaxoaguchi brigade^ the Ichiki 
reinforced battalion and other battalions 
of the 4ih and 124th Infantry Regiments 
(Nakaguma) already defeated or weak- 
ened by the Lunga defenders. 

By reason of the odd impasse in which 
both the Japanese and the Allied navies 
chose to avoid decisive battle to conserve 
their fleets, the Solomons waters changed 
hands every twelve hours, and thus each 
side kept an important trickle of aid going 
to its small combat force which represented 
the single point of ground contact between 
the belligerent powers. In daylight when 
Cactus could fly cover, the Allied ships 
came in from Espiritu Santo and other 
southern areas with reinforcements and 
supplies for the Marines. Barges, landing 
craft, and YP's shuttled errands across 
Sealark Channel. By nightfall the larger 
ships departed, and most of the others still 
in the Sealark area withdrew to safety in 
the Tulagi anchorage. Until dawn the 
elapanese took over. 

The destroyers and cruisers of the 
Tokyo Express habitually lurked in the 
Shortlands below Bougainville Island 
until the afternoon when they would start 
steaming south to be within 200 miles of 




Guadalcanal by about 1800. This was just 
inside the range of SBD's and TBF's from 
Henderson Field, but the maneuvering 
ships made poor targets, and the late hour 
gave the American planes time for only 
one crack at them before turning back for 
Lunga. After that the Express had an 
open line all the way to Sealark. 

While transport destroyers unloaded on 
either side of the Marine perimeter, Japa- 
nese warships stood close in at Lunga and 
went to work with their guns. Louie the 
Louse dropped flares to aid the naval gun- 
ners, and Washing-Machine Charlie 
lurked overhead to fritter out his bombs 
during lulls in surface firing. Under such 
attacks there was little the Marines could 
do but crouch in their foxholes and pray — 
or swear. Lunga defenders could estimate 
150 new enemy ground soldiers for every 
destroyer transport — often five or six a 
night — that made the Express run, and by 
early October these troops began to land 
insultingly close, just across the Matani- 
kau eight to ten miles from Henderson 
Field. The Allied turn to use the waters 
came at daylight, but U. S. forces did not 
have the man power to match the Japanese 
rate of reinforcement. 

Fortunately, the Japanese started 
slowly. Still thinking in terms of their 
operation against New Guinea, and mis- 
calculating Allied strength in the Solo- 
mons, Imperial planners only dribbled re- 
inforcements to Guadalcanal in August 
when the Marine position was particularly 
vulnerable. Not until after the Ichiki and 
Kawaguchi defeats did Japan begin to 
take serious stock of Vandegrift and his 

But now the Tokyo Express had stepped 
up its schedule, and by mid-October Hya- 
kutake had landed his 2d Division, two 

battalions of the 38th Division, one regi- 
ment and three batteries of heavy artillery, 
a battalion and a battery of mountain 
artillery, a mortar battalion, a tank com- 
pany, and three rapid-fire gun battalions. 
Special troops including engineers and 
medical personnel, and remnants of earlier 
attacks brought the Japanese force to 
about 20,000 men. 

Facing this mounting Japanese strength 
was a Marine force of about the same size. 
Arrival of the 7th Marines and the trans- 
fer of other troops from Tulagi bolstered 
General Vandegrif t's Lunga positions, but 
until 7 October there was little hope that 
more reinforcements would be forthcom- 
ing. Eear areas in the South Pacific had 
gained little strength since Vandegrift had 
argued for control of his 7th Marines, and 
the plan for the occupation of Ndeni still 
was in the pending basket. Marine 
strength thus promised to deteriorate 
while Japanese strength continued to 
mount. More than 800 Marine battle cas- 
ualties had been evacuated by early Octo- 
ber, and malaria continued to take its toll.^ 

The Cactus fliers were not doing much 
more than holding their own, either. By 
1 October, Lieutenant Colonel Mangrum's 
original VMSB-232 and Lieutenant Com- 
mander Caldwell's Flight 300 were done 
for, ^ Army pilots from the 67th Fighter 
Squadron had only about six or eight of 
their P^OO's in shape to fly, John Smith's 
VMF-223 had lost an even dozen pilots — 
six killed and six wounded — and other 
units, although stronger, still piled up 

* In October 1,960 malaria patients were hos- 

^ Mangrum was the only member of his outfit 
able to leave Henderson Field under his own 
power. He was evacuated on 12 October. Cald- 
well, who arrived at Lunga from the carrier 
Saratoga as a lieutenant, had been promoted. 



their share of losses. On the first day of 
October General Geiger had 58 planes; 
two days later the count stood at 49. 

If the Japanese had failed to win, place, 
or show with Ichiki, Kawaguchi, and Na- 
kaguma, the Allies likewise had been un- 
able to improve their odds by any comfort- 
able margin. To General Harmon the 
situation looked about as grim as it had on 
11 August when he expressed doubt that 
the Marines could hold their perimeter, 
and on 6 October he wrote to Admiral 
Ghormley that the Ndeni operation should 
be quashed until the situation improved. 
He questioned the logic of holding troops 
idle for a new operation when things were 
going so poorly in a battle already joined. 
He admitted certain factors favoring the 
Ndeni occupation, but he added that, ". . . 
in the final analysis they are individually 
or cumulatively vital to the success of main 
offensive operation or . . . maintaining 
security of South Pacific bases and lines of 
communications." ^ 

Specifically, Harmon recommended 
abandoning the Ndeni operation until the 
Guadalcanal situation improved; rein- 
forcements of Cactus (Guadalcanal) by at 
least one regimental combat team; the 
maximum possible intensification of naval 
surface action in South Solomons waters; 
and the prompt buildup of airdrome facili- 
ties and supplies at Henderson Field. 
Ghormley agreed that Vandegrift needed 
another regiment and that Henderson 
Field needed facilities and supplies, but 
the admiral retained for the time his plan 
to occupy Ndeni and build an airfield 
there. For the Guadalcanal reinforcement, 
Ghormley ordered Harmon to prepare a 
regiment of the New Caledonia garrison, 

'CGSoPac Itr to ComSoPac, 60ct42 (located 
at OCMH). 

and on 8 October he ordered Admiral 
Turner to embark the 164th Infantry of 
the Americal Division, Harmon's choice 
for the job, and depart Noumea for 
Guadalcanal on 9 October. 

It was to be a blockade run in force. 
Transports Zeilin and McCawley^ carry- 
ing supplies, 210 men of the 1st Marine 
Aircraft Wing and 85 Marine casuals as 
well as the 2,850 men of the Army regi- 
ment, sailed under escort of three de- 
stroyers and three mine layers while a 
larger force of four cruisers and five de- 
stroyers steamed off the convoy's left flank. 
These cruisers, San Francisco^ Salt Lake 
City^ Helena, and Boise and destroyers 
Buchanan, Dwrvean, Farenholt, Lajfey, and 
McCalla were commanded by Rear Ad- 
miral Norman Scott. Other U. S. Naval 
forces in the surrounding waters included 
Rear Admiral George D. Murray's Hornet 
carrier group some 180 miles southwest of 
Guadalcanal, and Rear Admiral Willis 
Augustus Lee's battleship Washington 
group about 50 miles east of Maliata. 
Scott's screening station for the unloading 
was near Rennell Island. 


On 11 October, while the ZeiJin and the 
McCawley made for their 13 October an- 
chorage schedule in Lunga Roads, Ad- 
miral Scott learned from aerial observers 
that two Japanese cruisers and six de- 
stroyers were bearing down The Slot. It 
was the night's Tokyo Express, Scott de- 
cided, and at 1600 he started toward 
Guadalcanal at 29 knots to intercept the 
run. His orders charged him to protect 
the transports, and to search for and de- 
stroy enemy ships and landing craft; he 
rushed eagerly to work. 



Actually Scott headed to intercept a 
force stronger than reports had indicated. 
Observers failed to spot three heavy 
cruisers, two seaplane carriers, and eight 
destroyers steaming some distance away 
-outside of The Slot. Japanese Vice Ad- 
miral Gunichi Mikawa, commander of the 
Eighth Fleet and the Outer Sea Forces^ 
and Vice Admiral Jinichi Kusaka, Elev- 
enth Air Fleet commander, had teamed up 
to strike the strongest blow yet against the 
bothersome Cactus fliers. In the after- 
noon of the 11th, Kusaka had 30 fighters 
and 35 bombers up to occupy Henderson 
fliers while Mikawa's bombardment and 
reinforcing groups steamed south outside 
the normal Japanese transport route. 
Heavy cruisers Aoha, Kinugasa^ and 
Furutaka with, destroyers Hatsuyvki and 
Fuhuki made up the bombardment group 
while the reinforcing fleet included sea- 
plane carriers Chitose and Nisshin, and de- 
stroyers AkiztiM, Asagumo, Natsugimi.o, 
Yam-agti/mo, Murakimw, and Shirayuki. 

By about 2200, while Scott maneuvered 
in the waters of Iron Bottom Sound be- 
tween Savo Island and Cape Esperance, 
the Japanese bombardment group came 
into The Slot and steamed south in a 
double column at 26 knots. At 2330 a 
spotting plane from USS San Francisco 
reported Japanese ships 16 miles from 
Savo and off Cape Esperance,* but Scott's 
ships still were unaware of the serious 
trouble facing them. Gunnery radar 
failed to pick up the enemy then approxi- 
mately 35 degrees forward of the port 
beam, and although the Helena earlier 
had spotted a Japanese ship bearing 315 
degrees and at a distance of 27,700 yards, 
she didn't report this contact for 15 

Flagship San Francisco, with rudimen- 
tary radar of that early period, had no 
contacts, and Scott continued to steam 
toward Savo with his ships in column. 
He counted this the best area for intercept- 
ing the Express he hoped to derail, and at 
about 2340 he had reversed course to head 
back toward the Cape when the Helena, 
at last confident about the blips from her 
better radar equipment, announced her fix 
of a target six miles away. Fortunately, 
since the U. S. fleet was having "eye" 
trouble, the Japanese ships were com- 
pletely blind, and even though certain 
communications misunderstandings ^ fur- 
ther delayed American fire, first salvos 
from the Helena at 2346 caught the enemy 
by complete surprise. Scott's ships had 
usurped Tokyo's turn in Sealark Channel. 

The Salt Lake City, Boise, and Faren- 
holt quickly added their fire to that of the 
Helena, and shortly thereafter the U. S. 
fleet crossed the Japanese "T" (sailed 
ahead of the Japanese column and at right 
angles to it) so that a majority of the 
American guns could bear on each Japa- 
nese ship as it came forward. The Japa- 
nese destroyer Fuhuki sank almost at once, 
the cruiser Fwrutaka took such a mauling 
that she limped away to sink later, and 
the Aoha caught fire. The only sound sur- 
vivors, cruiser Kinugasa and destroyer 
Hatsuyuki, withdrew. On the American 
side, the Boise, Salt Lake City, Farenholt, 
and Duncan suffered damage, and the 
Duncan sank the following day.® 

' These ships were from the reinforcement 

" For an account of these misunderstandings 
and for other descriptions of the Cape Esperance 
Battle see Struggle for Ouadalcanal, Chap VIII. 

'Also on 12 October Cactus fliers found the 
Japanese destroyers Murakumo and Natsugumo 
north of the Russell Islands, and their attack 
sank both of these ships. 



Scott could count the engagement a vic- 
tory, but it did not resolve the seesawing 
for power in the Solomons waters or skies. 
The Japanese only stepped up their air 
attacks on Henderson Field and continued 
preparations for the big push. 


Transports McCawley and Zeilen ar- 
rived at Kukum with the Army reinforce- 
ments early on 13 October, but this was 
one of the few bright spots of the day. 
Both radar and the Northern Solomons 
coastwatchers missed an air attack that 
came over at 1202, and the F4F's couldn't 
get up in time to liamper the 22 fighter- 
escorted bombers that rained down their 
bombs from 30,000 feet. Both Henderson 
Field and Fighter 1 were damaged, and 
fires from the attack burned 5,000 gallons 
of aviation fuel. 

Between 1330 and 1400 a second strike 
of 15 Japanese bombers caught most of 
the American planes back on their fields 
refueling. Some planes were damaged, 
and the strike undid the repair work that 
had been started by the 6th Seabees follow- 
ing the earlier raid. A few Cactus planes 
got up to pursue the Japanese, but the only 
American kill was scored by Captain Jo- 
seph J. Foss who had arrived on 9 October 
with Major Leonard K. Davis' VMF-121 
of MAG-14. The field was not completely 
out of action, but big bombers were ad- 
vised to avoid it except for emergencies. 

In spite of these interruptions. Colonel 
Bryant E. Moore managed to get his 164th 
Infantry ashore, along with other men 
and supplies from the transports, but 
trouble for the perimeter was not over. As 
the second bomber strike droned away, the 
150mm howitzers near Kokumbona were 
finally heard from. Safely beyond coun- 

terbattery range, these weapons began a 
slow methodical registration on the field 
and the perimeter. The fire was a brand 
of damage and destruction the men at 
Lunga had to live with, and so to have a 
pinpoint target for their anger if not their 
weapons they named this new entrant in 
their war Pistol Pete. 

Pete, as was most often the case with 
Louie the Louse and Washing-Machine 
Charlie, was plural. Hyakutake had 
landed 15 of these howitzers. But for the 
Marines and soldiers it was difficult to 
imagine batteries getting that personal, 
and Pete's particular brand of hell was a 
most personal and singular thing. So 
Pete became one enemy, the devil him- 
self — the devil and one big gun acting as 
Tojo's personal Nimrod. 

And after he thumped away at the 
perimeter all that day, an enemy task force 
built around battleships Haruna and 
Kongo came into Sealark Channel after 
nightfall to launch an 80-minute bombard- 
ment.'' This was the Japanese Combat 
Division 3, commanded by Vice Admiral 
Takeo Kurita, and it also included light 
cruiser Ismu and three ships of Destroyer 
Division 31 as a screen, plus a rear guard 
of four ships from Destroyer Division 16. 
The battleships had on board some new 
bombardment shells which had just ar- 
rived from the home islands. These had 
a greater bursting radius than former Jap- 
anese bombardment shells, and there were 
enough of them for battleships Haruna 
and Kongo to have 500 each. 

This was the first time that battleships 
had been used to bombard Henderson 
Field, and the Japanese hoped these big 
guns and the improved ammunition would 

' The unloaded American transports had de- 
parted late in the afternoon. 



completely knock out the Marine air and 
clear the way for a coordinated infantry 
attack. Louie the Louse illuminated the 
field, and the big guns cut loose. Coconut 
trees splintered, buildings and huts ripped 
open and crashed down, fragments and 
wreckage tore into planes and men, and 
more gasoline went up in bright fires which 
helped Japanese gunners stay on target for 
their systematic coverage of the field with 
more than 900 rounds of the high explosive 

As Admiral Tanaka described it later : 

The scene was topped off by flare bombs from 
our observation planes flying over the field, the 
whole spectacle making the Ryogoku fireworks 
display seem like mere child's play. The night's 
pitch dark was transformed by fire into the 
brightness of day. Spontaneous cries and shouts 
of excitement ran throughout our ships.* 

Then, as the ships became silent and 
withdrew east of Savo Island, the planes 
came back. Night bombers continued their 
strikes intermittently until daybreak, 
and by dawn of 14 October the Cactus Air 
Force could fly only 42 of the 90 planes 
that had been operational 24 hours earlier. 
Forty-one men had been killed and many 
more wounded, and the airfield was a com- 
plete shambles. Among the dead were 
Major Gordon A. Bell, whose VMSB-141 
had finally built up to 21 planes and fliers 
on 6 October, and four of his pilots : Cap- 
tains Edward F. Miller and Robert A. 
Abbott and Lieutenants Henry F. Chaney, 
Jr. and George L. Haley. 

Operations, sorely restricted by the loss 
of gasoline in the fire, moved to Fighter 1 
which was left in better condition than 
Henderson; and a few B-l7's which had 
been operating temporarily from Guadal- 
canal managed to bounce aloft from a 

' Tanaka Article, II, 815. 

2,000 yard stretch of Henderson that still 
was usable and fly back to Espiritu Santo. 
The Japanese "Pagoda," air headquarters 
since the early days, had been partially 
wrecked, and General Geiger had it bull- 
dozed away. It had proved too good a 
registration point for bombers, anyway. 

For the rest of the day the Japanese 
ships maintained their control of the 
waters around Guadalcanal, and planes 
continued to press their advantage in the 
air. Between the bombings and the shell- 
ings. Pistol Pete's effective interdiction 
prevented repair or use of the main air- 
strip, and by midafternoon Henderson had 
to be chalked off as completely unfit for 
use. By late afternoon fliers of the 
Army's 67th Fighter Squadron and 13 dive 
bomber pilots used Fighter 1 — and nearly 
all of Henderson's remaining supply of 
fuel — to strike back finally at the Japanese 
by attacking an early run of the Tokyo 
Express then only 70 miles north of 
Guadalcanal. One ship was sunk and an- 
other damaged, but the Express did not 
turn back. 

That night (14 October) the Japanese 
cruisers Chokai and Kinugasa moved down 
the channel to bombard Henderson Field 
while the express brought the six trans- 
ports carrying General Maruyama's 2d 
Division on down to Tassafaronga. The 
cruisers fired 752 eight-inch shells at the 
men around Lunga, and by dawn on 15 
October five of the enemy transports were 
clearly visible from the perimeter as they 
lay off Tassafaronga smugly unloading 
troops, supplies, and ammunition. 

Cactus fliers, smarting from the two-day 
hammering, drained gasoline from 
wrecked planes, searched the surrounding 
jungle for undamaged drums, and finally 
collected enough aviation fuel to mount an 
attack with the three SBD's that could still 



fly. But one of these planes had to be 
scratched when it tumbled into a crater 
on the way to the strip, and Lieutenant 
Robert M. Patterson lost SBD number two 
when the plane hit a shell hole while he 
raced for his takeoff. Patterson tried it 
again with the last dive bomber, and this 
time he made it. His single-plane attack 
did not hamper the Japanese much, but 
while he was flying, the ground crews 
quickly patclied other planes. It re- 
sembled an informal neighborhood boxkite 
club, with members hardly able to wait 
for work to be completed before they 
tested their craftsmanship. One at a time 
the first four planes were taken up to 
have a chance at the cocky Japanese trans- 
ports. Two minor hits were scored, but 
General Geiger stopped the assembly line 
combat action until he cauld muster more 

At 1000 Cactus was ready with 12 
SBD's, and they went up to drop 500- and 
1,000-pound bombs on the transports and 
then strafe their decks. That attack sank 
one of the transports. Next came attacks 
from P-39's and the relic P-400's, and fires 
broke out on two of the ships. After that, 
fliers from Espiritu Santo began to show 
up, and B-l7's and SBD's from the south 
sank another transport. The 'Tokyo Ex- 
press was in most serious trouble, in spite 
of 30 Zeros overhead to provide cover, 
and General Hyakutake might well have 
considered that the admirals and senior 
pilots in Rabaul had been somewhat over- 

" While this action was in progress, Army and 
Marine C-47's (R4D's) flew in with aviation 
gasoline, and seaplane tender MacFarland 
brought in additional supplies of the much- 
needed fuel. .Japanese planes next day (16 Oc- 
tober) damaged the tender, but she was repaired 
by her crew in an inlet of Florida Island. 

confident in this daring daylight delivery 
of his reinforcements. 

Even General Geiger's own pilot. Major 
Jack Cram, had his turn during that day 
of desperation when he made a run on the 
transports with two torpedoes slung under 
the wings of the general's Blue Goose, a 
bulbous and gouty PBY-5A. Cram got 
the torpedoes off, but then he was chased 
back to Fighter 1 by a clutch of Zeros, like 
sparrows around a ponderous hawk, and 
one determined enemy fighter had to be 
shot aAvay from the smoking Goose as 
Cram came in for his landing. 

By day's end three bombed transports 
of 7,000 to 8,000 tons each were beached 
and burning off Tassafaronga, and the 
other tw'o had fled back up Sealark Chan- 
nel and The Slot. But in spite of this, the 
Japanese had managed to unload 3,000 to 
4,000 men of the 330th and 16th Infantry 
Regiments as well as 80 per cent of the 
ships' cargo. These troops, the last the 
Japanese were able to land prior to their 
concentrated effort against the airfield, 
brought General Hyakutake's strength on 
the island to about 20,000 men. 

General Vandegrift now had approxi- 
mately 23,000 men, but the Marine force 
suffered severely from malnutrition, ma- 
laria, the exhaustive defensive actions, 
patrols, and field engineering work they 
had accomplished. Most of them were 
veterans, but in the unhealthy tropics that 
fact did not necessarily mean an advantage 
in the long run. Only the 164th Infantry 
of the Americal Division contained fresh 

With this additional regiment ashore, 
the division again reorganized the peri- 
meter, this time into five new defensive 
sectors. Clockwise from the Kukum area 
they were: Sector One — The 3d Defense 



Battalion with elements of the 1st Special 
Weapons Battalion, amphibian tractor- 
men, pioneers, and engineers who held 
7,100 yards of beach that straddled the 
Lunga River. ( See Map 23, Map Section) 

Sector Two— The 164th Infantry and 
elements of special weapons units with con- 
trol of a 6,600-yard line from the beach 
inland along the Ilu Eiver and thence west 
to a point near the east slope of Bloody 

Sector Three— The 7th Marines (less 3d 
Battalion), a 2,500-yard front of inland 
jungle from Bloody Ridge west to the 
Lunga River. 

Sector Four — The 1st Marines (less 3d 
Battalion), 3,500 yards of jungle from the 
Lunga west to the inland flank of the final 

Sector Five — The 5th Marines holding 
the northwest curve of the main perimeter 
from the flank of the 1st Marines north to 
the sea and then east along the beach to 
the west flank of the 3d Defense Battalion. 

Since the Japanese attack was expected 
fi'om the west across the Matanikau, the 
greatest strength was concentrated on that 
side of the perimeter. Forward of the 5th 
Marines' lines the 3d Battalions of both the 
1st and 7th Marines held a strong outpost 
line from the beach at the mouth of the 
river inland to Hill 67. This line was sup- 
ported by a battalion of the 11th Marines 
and elements of the 1st Special Weapons 
Battalion. The 3d Battalion, 2d Marines 
and 1st Tank Battalion units constituted 
the division reserve, and each regimental 
sector commander was directed to keep a 
third of his infantry strength in reserve 

Against these Marine and Army posi- 
tions. General Hyakutake prepared to 
launch his attack for the recapture of the 

airfield. On 15 October in Kokumbona he 
issued his attack order to Lieutenant Gen- 
eral Masao Maruyama's 2d Division. Date 
for the assault was set tentatively for 18 
October. The 2d Division would swing far 
inland to hit the Marines from the south 
with a night attack in two columns of bat- 
talions while the Seventeenth Army artil- 
lery commander. General Sumiyoshi, 
would shell the perimeter and then launch 
a diversionary strike with infantry units 
near the mouth of the Matanikau. For 
this coastal attack Sumiyoshi had a force 
of some 2,900 men comprising the bat- 
talions of the J(.th Infantry plus a tank 
company, seven light field artillery pieces, 
fifteen of the 150mm howitzers, and three 
100mm guns. 

For his inland attack, Maruyama had 
some eight or nine infantry battalions 
totaling 5,600 men, plus artillery and 
supporting troops. General Kawaguchi, 
who had tried his hand in the same area 
before, would command the right arm of 
the assault with two battalions of the 230th 
Infantry, one battalion of the In- 
fantry, and elements of the 3d Light 
Trench Mortar Battalion, 6th and 7th In- 
dependent Rapid Gun Battalions, the 20th 
Independent Mountain Artillery, engi- 
neers, and medical troops. The left attack- 
ing column would be under command of 
Major General Yumio Nasu and would in- 
clude the 29th Infantry, the remainder of 
the 3d Light Trench Mortar Battalion, a 
rapid fire gun battalion, a mountain artil- 
lery battalion, and engineers. The 16th 
Infantry and some engineers — a part of 
Nasu's command — would be in reserve be- 
hind the 29th Infantry. 

General Hyakutake was confident of 
success. He had left the bulk of his 38th 
Division at Rabaul. Banzai was to be 



Maruyama's signal of victory at the air- 
field, and his attack from the south was 
ordered to press unrelenting destruction 
upon the enemy until General Vandegrift 
himself came forth to surrender. 

Thus charged, General Maruyama 
struck out through the jungle wilderness 
on 16 October. 


Transportation was pedestrian, cargo 
moved on bended backs, and hand power 
drove the engineering tools. Thus the 
column of enveloping Japanese inched 
single file across the tortuous Guadalcanal 
back country like a segmented serpent 
crawling through the perpetual wet shad- 
ows of the tropical forest. 

The so-called Maruyama Trail, begun 
by engineers in September, scratched its 
thin scar along the floor of the jungle 
southward from Kokumbona, east across 
the Matanikau and the Lunga inland from 
Mount Austen, and then north to an assem- 
bly area south of Bloody Eidge. Safely 
beyond range of Marine patrols and 
hidden from aerial view by the vine-laced 
tops of giant hardwoods, the Japanese 
soldier moved with an artillery or mortar 
shell lashed to his already heavy load of 
normal equipment, frequently used ropes 
to scale the rough ridges and steep valleys, 
and by turns tugged a line or hunched his 
shoulder to the common effort of manhan- 
dling artillery, mortars, and machine guns. 

Heavy rain fell almost every day. The 
van of the single-file advance often had 
completed its day's march and bivouacked 
for the night before the rear elements were 
able to move. Troops weakened on their 
half ration of rice. Heavy artillery pieces 
had to be abandoned along the route, and 

mortars also became too burdensome to 
manage. Frequently unsure of their exact 
location in the jungle, the Japanese by 19 
October still had not crossed the upper 
Lunga, and Maruyama postponed his 
assault until the 22d. Meanwhile Gen- 
eral Sumiyoshi's fifteen Pistol Petes 
pounded the Lunga perimeter, air attacks 
continued, and Imperial warships steamed 
brazenly into Sealark Channel nearly 
every night to shell the airfield, beaches, 
and Marine positions. 

The tempo of action obviously was build- 
ing up for the counteroffensive, and 
Marines and soldiers worked constantly to 
improve their field fortifications and keep 
up an aggressive patrol schedule. Patrols 
did not go far enough afield, however, to 
discover Maruyama's wide-swinging en- 
veloping force, and reconnaissance to the 
east found no indications of a Japanese 
build-up on that flank. Thus General 
Vandegrift and his staff were aware only 
of Sumiyoshi's threat along the coast from 
the west. 

There the first probe came on 20 October. 
A Japanese combat patrol, augmented by 
two tanks, ventured into view on the west 
bank of the Matanikau but turned back 
after one tank was knocked out by 37mm 
fire from the lines of the 3d Battalion, 1st 
Marines. Sporadic artillery fire was the 
only Japanese answer to this checkmate, 
and it continued until sunset the next day. 
Then the artillery fire intensified briefly, 
and nine infantry-supported tanks de- 
bouched- from the west bank jungle and 
drove eastward for the sandspit at the 
mouth of the river. But again the fire 
from a 37mm stopped one of the tanks, 
and the attack turned back without seri- 
ously threatening the river-mouth posi- 
tions of Company I, 3/1. The Marine 



FIVE BLASTED JAPANESE TANKS knocked out by Marine 37mm guns during the abortive attempt to 
force the perimeter along the mouth of the Matanikau. (USMC 54898) 

MARINE LIGHT TANKS, mounting machine guns and 37mm cannon, were severely hampered in their 
operations by the jungle terrain of Guadalcanal. (USN 18525 ) 



battalion had taken a few casualties from 
artillery and mortar fire, but neither of 
these first two attacks had posed a serious 

At the Matanikau positions on 22 Octo- 
ber Sumiyoshi continued firing his mor- 
tars and artillery but mounted no new 
assault. Inland, General Maruyama 
struggled with the jungle some distance 
from his lines of departure, and he was 
forced to postpone his proposed assault to 
23 October. But on that day he still was 
unprepared to attack and again he set back 
his plans another 24 hours. 

At about 1800 on the 23d, however, 
Sumiyoshi once more intensified his artil- 
lery and mortar fire to lay down an ortho- 
dox preparation pattern on the Marine 
east bank positions and along the coastal 
route from the Lunga Perimeter. Near 
the end of evening nautical twilight the 
artillery fire ceased, and a column of nine 
18-ton medium tanks churned across the 
sandspit in an attempt to force a penetra- 
tion. In assembly areas to the rear in- 
fantry troops stood by to assault in the 
wake of the tanks. 

Slim-barreled 37's again blasted at the 
Japanese tanks while infantry mortars 
and howitzers of the 11th Marines dumped 
prearranged concentrations farther west 
to break up the pending infantry assault. 
The enemy ground troops never got 
started, and the tank charge miscarried 
when eight of the vehicles were hammered 
to a standstill by the 37's. One tank man- 
aged the crossing but staggered out of 
control when a Marine pitched a grenade 
in its track as it lumbered by his foxhole. 
Pursued by a half-track 75, the beset ma- 
chine wallowed into the surf where it 
stalled to form a sitting duck target for 
the tank destroyer. 

The other eight hulks remained strewn 
along the sand bar across the river mouth, 
and artillery fire knocked out three more 
tanks that never got to attack. Hundreds 
of the enemy soldiers who had been wait- 
ing to follow the tanks were killed. The 
action was over by 2200, although at about 
midnight the Japanese made a half- 
hearted attempt to cross the river farther 
upstream. This thrust was turned back 
with little trouble. 

From his study of interrogations of the 
Japanese generals involved. Dr. John Mil- 
ler, Jr., sums up : 

Sumiyoshi had sent one tank company and 
one infantry regiment forward to attack a pre- 
pared position over an ohvious approach route 
while the Americans were otherwise unengaged. 
The Maruyama force, still moving inland, had not 
reached its line of departure. In 1946, the re- 
sponsible commanders gave different reasons for 
the lack of co-ordination and blamed each other. 
According to Hyakutake, this piecemeal attack 
had been a mistake. The coastal attack was to 
have been delivered at the same <ime as Maru- 
yama's forces struck against the southern perime- 
ter line. Maruyama, according to Hyakutake, 
was to have notified the 4th Infantry when he 
reached his line of departure on 23 October, and 
he so notified the 4th Infantry. The regiment 
then proceeded with its attack. 

Maruyama disclaimed responsibility for the 
blunder, and blamed 17th Army Headquarters. 
His forces, delayed in their difiicult march, had 
not reached their line of departure on 23 October. 
The 17th Army, he asserted, overestimated the 
rate of progress on the south fiank and ordered 
the coast forces to attack on 23 October to guar- 
antee success on the south flank. 

Sumiyoshi was vague. He claimed that 
throughout the counteroffensive he had been so 
weakened by malaria that he found it difiicult 
to make decisions. Despite an earlier statement 
that he did not know why the attack of 23 October 
had been ordered, he declared that he had at- 
tacked ahead of Maruyama to divert the Amer- 
icans. Communication between the two forces, 
he claimed, had been very poor. Radio sets gave 
off too much light, and thus had been used only 



ill daylight hours. Telephone communication 
had been frequently disrupted. As a result the 
coast force had been one day behind in its 
knowledge of Maruyama's movement.*" 

Meanwhile the Marine division " had 
started a shift of manpower within the 
perimeter. In the face of Sumiyoshi's at- 
tacks, and with no patrol contacts to the 
south or east, the 2d Battalion, 7th Marines 
on 23 October pulled out of its southern 
lines east of the Lunga and moved west 
to relieve the 3d Battalion, 1st Marines at 
the mouth of the Matanikau. This left the 
1st Battalion, 7th Marines (Puller) with a 
responsibility for the defense of all of 
Sector Five, the 2,500-yard defense line 
from the inland flank of the 164th Infan- 
try west across the southern slopes of 
Bloody Ridge to the Lunga River. Pul- 
ler's extended lines were thin, but there 
appeared very little danger from the south. 

Hanneken's 2/7 did not effect its in- 
tended relief, however, because of the 
heavy Japanese artillery fire that engaged 
3/1 on the 23d, and on the following day a 
new assignment was given to the 7th Ma- 
rines battalion. On the 24:th the Marines 
of 3/7 on Hill 67 south of the Matanikau 
mouth had spotted a Japanese column, ob- 
viously a flanking force,^^ moving east 
across Mount Austen's foothills. Artillery 
and air was called in on this enemy move- 
ment, but the Japanese disappeared into 

" Miller, Guadalcanal, 157-159, quoted by per- 
mission of the author. 

" BriGen Rupertus, ADC, became acting CG 
of the 1st MarDiv on 23 October. MajGen Vande- 
grift left at dawn that day for conferences at 
Noumea, flying out with LtGen Thomas Holcomb, 
Marine Corps Commandant, whose Pacific tour 
had brought him to Guadalcanal on 21 October. 

"This force, never positively identified in re- 
constructions of battle events, is thought to have 
been that of Col Oka which appears later in night 
attacks of 25-26 October. FinalRept, Phase V, 22. 

jungle ravines about 1,000 yards south of 
Hill 67 before they could be engaged. In 
the face of this threat apparently headed 
for tlie 4,000-yard gap between the Matani- 
kau outpost and the Lunga perimeter, 2/7 
was assigned to plug this hole, and the 3d 
Battalion, 1st Marines retained its posi- 
tions overlooking the beach and the Ma- 

Later the same day came other indica- 
tions that the Sumiyoshi action would not 
be the only Japanese effort against the 
perimeter. Late in the afternoon of 24 
October an observer in the 1/7 lines south 
of the airfield saw a Japanese officer study- 
ing Bloody Ridge through field glasses, 
and a scout-sniper patrol reported seeing 
the smoke from "many rice fires" in the 
Lunga valley about two miles south of 
Puller's positions on the Ridge. By this 
time twilight was settling over Guadal- 
canal, and there was little the Marines 
could do but wait out developments from 
existing positions. The only troops not in 
front lines were those in reserve in the 
various defensive sectors and the 3d Batta- 
lion, 2d Marines, the division reserve, then 
bivouacked north of Henderson Field. 

The rice fires and the officer with field 
glasses undoubtedly were signs — and the 
first the Marines had — of the reinforced 
2d Division that finally had negotiated 
the grueling advance from Kokumbona 
over the Maruyama Trail. With all his 
artillery and mortars strewn along the 
route behind him, Maruyama at last had 
crossed the Lunga into his assembly areas 
south of Bloody Ridge. There the force 
stood at twilight on 24 October ready to 
attack with only infantry weapons against 
the dug-in Marines who were backed up 
by artillery and mortars. 

Hoping for bright moonlight to aid co- 
ordination (the night actually went black 



with heavy rain), the Japanese general 
ordered a narrow attack over the ground 
Kawaguchi's force had assaulted in mid- 
September. The Tnain effort was assigned 
to the 29th Infantry, with the 16th Inf cm- 
try in reserve, while farther to the east the 
Kawaguchi command — now led by Colonel 
Toshinari Shoji ^^ — was to make a paral- 
lel assault. 

At about 2130 a Japanese unit clashed 
briefly with a 46-man outpost Puller had 
stationed forward of his tactical wire, but 
after a short fire fight the enemy bypassed 
the position, and the battlefield was quiet. 
Platoon Sergeant Ralph Briggs, Jr., in 
charge of the outpost, notified Puller that 
a large force of Japanese were moving 
about the outpost hill toward the battalion 
lines, but Puller ordered his men to hold 
fire so that Briggs could infiltrate to safety. 
But the outpost already was flanked by the 
Japanese moving around the hill, and 
Briggs led his men to the east while the 
enemy moved closer to Puller's battalion 
and began to cut the tactical wire in front 
of the 1/7 positions." 

While Puller's men strained to hear the 
approaching enemy above the sound of 
drumming rain which lashed the night, the 
Japanese prepared their routes through 
the Marine barbed wire and formed up for 

" Gen Kawaguchi, possibly with a justifiable 
dislike for this ridge terrain, had advocated an 
attack farther to the southeast, had thereby 
fallen from favor and had been relieved by 
Maruyama. Miller, Guadalcanal, citing Sumi- 
yoshi and Tamaki (2d Div CofS'), 160. 

" Thirty-three members of this outpost man- 
aged to reach the lines of the 164th Inf the next 
day, but 13 men remained lost and hunted by the 
Japanese. Nine of these finally returned to safety 
after many harrowing adventures with the 
jungle and enemy, although one of the nine was 
gone for two weeks. Four of the wanderers were 
killed by the Japanese. 

their attack. Then at 0030 on 2'5 October, 
Nasu 's men came out of the jungle scream- 
ing their hanzais, throwing grenades, and 
firing rifles and machine guns to strike the 
left center of 1/7's fine with an assault 
in depth on a narrow front. Puller called 
in mortar and artillery concentrations, his 
riflemen took up a steady fire, and the ma- 
chine guns rattled almost endless bursts 
down their final protective lines. 

From Puller's left, troops of the 2d 
Battalion, 164th Infantry added their fire 
to that of the Marines, but still the Japa- 
nese assaulted, trying to rush across the 
fields of fire toward the Ridge. The at- 
tack kept up for 10 or 15 minutes, but 
finally ground itself to a halt against the 
combined arms of the U. S. force. Then 
there was a lull while the Japanese re- 
grouped and came back again, trying to 
clear a penetration with their grenades and 
small arms. The Marine commander as- 
sessed correctly that his men were stand- 
ing off the main attack of Rabaul's big 
counteroffensive, and that the force in the 
jungle to his front obviously was strong 
enough to keep such attacks going most 
of the night. He called for reinforce- 
ments, and division headquarters ordered 
Lieutenant Colonel Robert K. Hall to take 
his 3d Battalion of the 164th Infantry 
down the Ridge to bolster Puller's thin 

But the reinforcements had a mile of 
muddy ridge to cover before they could be 
of any help, and in the meantime the Ja- 
panese continued to assault out of the jun- 
gle and up the slopes. A small group 
forced a salient in the Marine line to fall 
upon a mortar position, and farther to the 
front Nasu's soldiers worked close to a 
water-cooled machine gun and knocked 
out all but two of its crew. Marines near 



the mortar position won back the tube from 
the enemy, and in the machine-gun section 
Sergeant John Basilone took rescue mat- 
ters into his own hands. For this action 
and later heroism in braving Japanese fire 
to bring up ammunition, Basilone became 
the first enlisted Marine of World War II 
to win the Medal of Honor. ^^ 

As these attacks continued, Colonel 
Hall's soldiers began to arrive in small de- 
tachments. Puller made no attempt to 
give this battalion a line of its own on his 
threatened front, but instead had his men 
lead these fresh troops into his line where 
they were most needed at the moment. The 
fighting was too brisk and the night too 
rainy for any major reshuffling of lines. 
By 0330 the reinforcement was complete, 
and the Japanese attacks were becoming 
less intense. Infantry and supporting 
fires had cut down the Nasu force so that 
each new assault was made with fewer and 
fewer men. 

Fortunately, all had not gone well for 
the Japanese plans. Nasu bore the brunt 
of the effort without assistance to his right 
where the second assaulting column was 
to have struck. Colonel Shoji, with Kawa- 
guchi's former command, had strayed out 
of position in the difficult terrain and poor 
weather and got in behind General Nasu's 
29th Infantry. Sho j i was unable to correct 
this error in time for his battalions to par- 
ticipate in the action. 

But Maruyama was true to his orders to 
press unrelenting attacks upon the Ameri- 
cans. With characteristic resolution, the 
Japanese struck at the Marines again and 
again throughout the night. The Bushido 
spirit was unswerving, but the flesh could 
not endure the concentrated fire from the 

'° Basilone was killed in 1945 during the Marine 
assault of Iwo Jima. 

combined U. S. infantry battalions, the 
artillery, and 37mm's from the neighbor- 
ing 2d Battalion, 164th infantry. By 
dawn Maruyama called back his men to re- 
group for later attacks, and Puller and 
Hall began to reorganize their intermin- 
gled battalions and readjust their lines. 
The first strong effort of the counteroffen- 
sive had been turned back, but the re- 
mainder of 25 October, Sunday in the 
Solomons, was not a restful day. 

Heavy rains on the 23d and 24th had 
turned Fighter 1 into a mud bog, and at 
0800 Pistol Pete opened up again on 
Henderson to fire at ten-minute intervals 
until 1100. With Cactus fliers thus effec- 
tively grounded, enemy planes from 
Rabaul took advantage of this, and the 
first fair weather in three days, by attempt- 
ing to give the Japanese counteroffensive 
some semblance of the coordination that 
Generals Sumiyoshi and Maruyama had 
muffed. Likewise strong enemy naval 
forces, to be engaged next day in the Bat- 
tle of Santa Cruz, were known to be ap- 
proaching, and early in the morning three 
Japanese destroyers, as bold as the Zeros 
overhead, cavorted into Sealark Channel 
to chase off two American destroyer-trans- 
ports, sink a tug, set fire to two harbor 
patrol craft, and harass the beach positions 
of the 3d Defense Battalion. Finally ven- 
turing too close to shore, one of the enemy 
destroyers was chastised by three hits 
from 5-inch guns of the defense battalion, 
and the Japanese ships then withdrew. In 
all, the day earned its name of "Dugout 

But the name "was a misnomer in a 
sense." ^^ Although the lurking Zeros kept 
"Condition Red" alerts in effect most of 
the day, bombing raids came over only 

The Island, 178. 



twice,^'^ and Lunga defenders not connected 
with Cactus operations climbed out of their 
foxholes to watch the dogfights which be- 
gan after Fighter 1 dried enough to sup- 
port takeoff s. These American planes 
were able to go up at 1430 to meet a 16- 
bomber strike from Rabaul and hamper 
this attack ; and a nine-plane bombing raid 
at 1500 dumped its